Hazinfo Technical Report

Evaluating Last-Mile Hazard Information
Dissemination: A Research Proposal
(HazInfo Project)
Grant No.: 103553-001
Grant Period: 01-December-2006 to 31-March-2008
Report Presentation Date: 10-March-2008
Submitted to IDRC by:

LIRNEasia
12 Balcombe Place
Colombo 8 Sri Lanka
Tel: +94 114 939 992 Fax: +94 112 675 212
E-mail: samarajiva@lirne.net
Web site: www.lirneasia.net
SRI LANKA
LIRNEasia is regional, non-

Grant No. 103553-001 Report Submitted: March 01, 2008
TABLE OF CONTENT
LIST OF TABLES...................................................................................................................................8
LIST OF FIGURES...............................................................................................................................11
1.SYNTHESIS........................................................................................................................................14
2.RESEARCH PROBLEM...................................................................................................................15
2.1PROJECT OBJECTIVE............................................................................................................................................................15
2.2PROJECT BACKGROUND........................................................................................................................................................15
2.3COMMENTS ON THE PROPOSED RESEARCH...............................................................................................................................16
3.RESEARCH FINDING......................................................................................................................17
3.1.PERFORMANCE OF THE ICTS AS A WARNING TECHNOLOGY......................................................................................................18

3.2.CONTRIBUTION OF THE TRAINING REGIME..............................................................................................................................20
3.3.CONTRIBUTION OF VILLAGE ORGANIZATIONAL DEVELOPMENT..................................................................................................27
3.4.GENDER SPECIFIC RESPONSE TO HAZARD MITIGATION ACTION................................................................................................29
3.5.DEGREE OF INTEGRATION OF ICTS IN THE DAILY LIFE OF VILLAGES...........................................................................................29
3.6.LOCAL BARRIERS IN RESPONDING TO ALERTS...........................................................................................................................30
4.FULLFILLMENT OF PROJECT OBJECTIVES..........................................................................32
5.PROJECT DESIGN AND IMPLEMENTATION..........................................................................36
5.1.CONCEPT OF OPERATIONS....................................................................................................................................................36
5.2.LM-HWS FIVE WIRELESS ICTS..........................................................................................................................................37
5.3.ICT PARTNERS AND THEIR CONTRIBUTIONS...........................................................................................................................39
5.4.CAP GUIS AND CAP BROKER..........................................................................................................................................40
6.PROJECT OUTPUT AND DISSEMINATION..............................................................................41
6.1.INFORMATION SHARING AND DISSEMINATIONS........................................................................................................................41

6.2.KNOWLEDGE CREATION......................................................................................................................................................45
6.3.TRAINING..........................................................................................................................................................................45
7.CAPACITY BUILDING....................................................................................................................48
7.1.LOCAL RESEARCH AND DEVELOPMENT..................................................................................................................................48

7.2.HAZARD INFORMATION HUB MONITOR COURSE.....................................................................................................................48
7.3.VILLAGE TRAINING OF COMMUNITY FIRST-RESPONDERS..........................................................................................................48
8.PROJECT MANAGEMENT............................................................................................................50
9.IMPACT..............................................................................................................................................51
10. OVERALL ASSESSMENT............................................................................................................52
11.RECOMMENDATIONS.................................................................................................................53
11.1.MESSAGE FORMATTING - THE HIH AND THE GOVERNMENT....................................................................................................53

11.2. FURTHER TEST ICT ASSESSMENT TOOLS AND METHODOLOGIES...........................................................................................53
11.3. TABLE TOP EXERCISE FOR SELECTED MEMBERS IN COMMUNITY...........................................................................................53
11.4. TRAINING AND CERTIFICATION OF HIH MONITORS..............................................................................................................54
11.5. TRAINING OF ICT GUARDIANS.........................................................................................................................................54
11.6.COMMUNITY-BASED EMERGENCY COMMUNICATIONS PLANNING AND TRAINING.........................................................................54
11.7.CONTRIBUTION TO CAP RESEARCH THROUGH WORKING GROUPS..........................................................................................55
11.8.ENHANCE AND TEST ICTS WITH COMPLETE AND FULL CAP FEATURES..................................................................................55
11.9.FREE AND OPEN SOURCE SOFTWARE CAP BROKER.............................................................................................................56
11.10.CLOSED USER GROUP DIGITAL AUDIO BROADCASTING.......................................................................................................57
12.TECHNICAL ANNEX 1: LIST ACRONYMS AND ABBREVIATIONS.................................59
13.TECHNICAL ANNEX 2: WHAT IS A COMMUNITY-BASED LM-HWS..............................61
14.TECHNICAL ANNEX 3: INTERNETWORKING PIPES, AND GATEWAYS......................62
15.TECHNICAL ANNEX 4: RESEARCH MATRIX AND ICT ALLOCATIONS.......................66
16.TECHNICAL ANNEX 5: CALCULATING THE EFFICIENCY OF PROCESSES...............67
17.TECHNICAL ANNEX 6: MEASURING EFFECTIVENESS OF THE HIH AND

COMMUNITY.......................................................................................................................................70
18.TECHNICAL ANNEX 7: CALCULATING RELIABILITY OF ICT.......................................70
18.1.CERTAINTY OF MESSAGE RECEIPT......................................................................................................................................71

18.2.EFFICIENCY MEASURE OF TRANSMISSION.............................................................................................................................73
19.TECHNICAL ANNEX 8: MEASURING EFFECTIVENESS OF THE ICT............................73
19.1.EFFECTIVENESS OF THE INDIVIDUAL ICTS............................................................................................................................73

19.2.EFFECTIVENESS OF A COMBINATION OF ICTS.......................................................................................................................74
20.TECHNICAL ANNEX 9: FUZZIFICATION OF FULL CAP COMPLETENESS..................74

20.1.CAP PROFILE FOR SRI LANKA..........................................................................................................................................74

20.2.COMPULSORY ELEMENTS OF THE CAP PROFILE...................................................................................................................75
20.3.LANGUAGE DIVERSITY IN SRI LANKA.................................................................................................................................75
20.4.MIX OF PICTORIAL, AUDIO AND TEXT COMMUNICATION MEDIUMS ........................................................................................76
21.TECHNICAL ANNEX 10: FUZZIFICATION OF ALERTING PROCESS.............................77
21.1.ACKNOWLEDGEMENT OF MESSAGE RECEIPT.........................................................................................................................77

21.2.WAKEUP FUNCTION TO GAIN ATTENTION..............................................................................................................................78
22.TECHNICAL ANNEX 11: FUZZIFICATION OF TWO-WAY COMMUNICATION...........79
22.1.BI-DIRECTIONALITY FOR UPSTREAM AND DOWNSTREAM COMMUNICATION...............................................................................79
................................................................................................................................................................80
23.TECHNICAL ANNEX 12: FUZZIFICATION OF ADOPTABILITY.......................................80
23.1.INTEGRATION OF ICT IN TO COMMUNITY DAILY LIFE.............................................................................................................80

23.2.TOTAL COST OF OWNERSHIP..............................................................................................................................................80
24.TECHNICAL ANNEX 13: FUZZIFICATION OF MINIATURERIZATION..........................82
24.1.WEIGHT OF THE ENTIRE FUNCTIONAL UNIT............................................................................................................................83

24.2.VOLUME OF TERMINAL DEVICE AND PERIPHERALS..................................................................................................................83
24.3.LONGEVITY OF OPERATIONAL STATE....................................................................................................................................85
25.TECHNICAL ANNEX 14: EXPECTED PERFORMANCE OF THE ICT...............................86
25.1.WORLDSPACE ADDRESSABLE SATELLITE RADIO SYSTEM......................................................................................................86

25.2.DIALOG DISASTER AND EMERGENCY WARNING SYSTEM........................................................................................................89
25.3.SOLANA NETWORKS/INNOVATIVE TECHNOLOGIES VSAT/IPAS............................................................................................94
25.4.SRI LANKA TELECOM NOMADIC TELEPHONY SYSTEM...........................................................................................................96
26.TECHNICAL ANNEX 15: GUIDELINES FOR THE HIH.........................................................99
26.1.OVERVIEW......................................................................................................................................................................99
26.2.THE LM-HWS AND ITS RESPONSIBILITIES FOR ISSUING ALERTS..........................................................................................100
26.3.STAFFING AND TRAINING AT THE HIH..............................................................................................................................102
26.4.MONITORING.................................................................................................................................................................107
26.5.ACTIVATION..................................................................................................................................................................108
26.6.PROHIBITED PRACTICES...................................................................................................................................................121
26.7.TESTING........................................................................................................................................................................122
26.8.ANNEX A: CONTACT INFORMATION..................................................................................................................................123
26.9.ANNEX B: GUIDELINES FOR DOCUMENTING AN EOI...........................................................................................................123
26.10.ANNEX C: SECONDARY SOURCES FOR CONFIRMING EVENTS OF INTEREST..............................................................................124
26.11.ANNEX D: TELEPHONE “HOTLINE” NUMBER FOR CONTACTING THE HIH..............................................................................125
26.12.ANNEX F: CAP ALERT MESSAGE EXAMPLE......................................................................................................................126
27.TECHNICAL ANNEX 16: TRAINING – HAZARDS...............................................................127
27.1.DEFINITION OF HAZARDS.................................................................................................................................................127
27.2.HAZARD EVENTS...........................................................................................................................................................128
27.3.MULTIPLE HAZARDS.......................................................................................................................................................128
27.4.THE RETURN PERIOD......................................................................................................................................................128
27.5.CLASSIFICATION OF HAZARDS..........................................................................................................................................129
27.6.MORE HAZARD TERMINOLOGY........................................................................................................................................130
27.7.HAZARD IDENTIFICATION AND ASSESSMENT.......................................................................................................................131
28.TECHNICAL ANNEX 17: TRAINING – VULNERABILITY AND RISK.............................137
28.1.VULNERABILITY.............................................................................................................................................................137
28.2. RISK............................................................................................................................................................................139
29.TECHNICAL ANNEX 18: TRAINING – DISASTER MANAGEMENT................................145
29.1. DISASTER MANAGEMENT................................................................................................................................................145

29.2. RISK MANAGEMENT......................................................................................................................................................146
29.3. PREVENTION.................................................................................................................................................................148
29.4.CRISIS MANAGEMENT.....................................................................................................................................................148
29.5.RESPONSE AND RELIEF....................................................................................................................................................149
29.6.REHABILITATION AND RECONSTRUCTION............................................................................................................................149
30.TECHNICAL ANNEX 19: TRAINING – PARTICIPATORY RURAL APPRAISAL TOOLS

...............................................................................................................................................................150
30.1.IMPORTANCE OF PARTICIPATORY TECHNIQUES....................................................................................................................150

30.2.PARTICIPATORY RURAL APPRAISAL...................................................................................................................................151
30.3.AIMS OF PARTICIPATORY METHODS....................................................................................................................................151
30.4.SCOPE OF PRA.............................................................................................................................................................151
30.5.PARTICIPATORY COMMUNITY MONITORING AND EVALUATION.................................................................................................153
31.TECHNICAL ANNEX 20: TRAINING – COMMUNITY PUBLIC AWARENESS..............155
31.1.DEFINITION OF SOCIAL CHANGE COMMUNICATION CAMPAIGN..............................................................................................155

31.2.DEFINITION OF DISASTER RISK COMMUNICATION................................................................................................................156
31.3.STEPS IN DRC..............................................................................................................................................................159
31.4.BARRIERS TO MAKING CHANGE.......................................................................................................................................159
31.5.SOME DISTINCTIONS.......................................................................................................................................................160
31.6.PARTNERSHIPS...............................................................................................................................................................160
31.7.DRC BUDGET...............................................................................................................................................................161
32.TECHNICAL ANNEX 21: TRAINING – EARLY WARNING................................................162
32.1.CLASSIFICATION OF DISASTERS FOR EARLY WARNING ACTIVITIES............................................................................................163

32.2.WARNING SYSTEMS........................................................................................................................................................163
32.3.IMPORTANT ISSUES.........................................................................................................................................................164
32.4.EVACUATION PREPAREDNESS RESPONSIBILITY.....................................................................................................................165
33.TECHNICAL ANNEX 22: TRAINING – RESPONSE PLANNING.......................................166
33.1.INTRODUCTION...............................................................................................................................................................166
33.2.WHAT TO KNOW? WHAT TO PLAN..................................................................................................................................167
33.3.EVACUATION POLICY AND PROCEDURE..............................................................................................................................167
33.4.HOUSE HOLD LEVEL PREPARATION..................................................................................................................................168
IF WARNING TIME IS ADEQUATE.................................................................................................................................................169
33.5.TRAINING AND DRILLS....................................................................................................................................................169
34.TECHNICAL ANNEX 23: TRAINING –MONITORING AND EVALUATION...................171
34.1.DEFINITION OF MONITORING AND EVALUATION..................................................................................................................171

34.2.PLACE OF MONITORING AND EVALUATION IN DRC............................................................................................................172
34.3.MONITORING PLAN AND MECHANISM................................................................................................................................173
34.4.EVALUATION TERMINOLOGY............................................................................................................................................173
34.5.EVALUATION PLAN.........................................................................................................................................................174
34.6.OUTCOME EVALUATION DESIGNS.....................................................................................................................................175
34.7.COMMUNITY INDICATORS.................................................................................................................................................175
34.8.STEPS IN EVALUATION RESEARCH.....................................................................................................................................175
34.9.DOCUMENTATION...........................................................................................................................................................176
35.APPENDIX A: MAP OF SRI LANKA WITH 32 COASTAL COMMUNITIES....................177

36.APPENDIX B: DATA OF RESEARCH MATRIX ALLOCATION........................................178
37.APPENDIX C: DATA OF HIH OPERTIONAL.........................................................................179
38.APPENDIX D: DATA OF EFFICIENCY MEASURE FROM LIVE-EXERCISES..............182
39.APPENDIX E: DATA FOR CERTAINTY MEAURE OF ICT................................................187
40.APPENDIX F: EFFECTIVENESS DATA FOR ALL THE CLIQUES...................................190
40.1.FULL CAP COMPLETENESS.............................................................................................................................................190

40.2.ACTIVE ALERT FUNCTION...............................................................................................................................................191
40.3.ADOPTABILITY...............................................................................................................................................................191
40.4.MINIATURIZATION..........................................................................................................................................................193
40.5.TWO-WAY.....................................................................................................................................................................194
41.APPENDIX G: OUTCOME OF THE ICT IN THE FIELD......................................................195
41.1.WORLDSPACE ADDRESSABLE SATELLITE RADIO SYSTEM....................................................................................................195

41.2.DIALOG DISASTER AND EMERGENCY WARNING SYSTEM......................................................................................................197
41.3.SOLANA NETWORKS/INNOVATIVE TECHNOLOGIES VSAT/IPAS..........................................................................................199
..............................................................................................................................................................201
42.APPENDIX H – DATA FROM SILENT TESTS........................................................................202
43.APPENDIX I – OUTCOME OF ICT ON 12-SEP-2007 TSUNAMI ADVISORY...................209
44.APPENDIX J – DATA COLLECTION SHEETS......................................................................211
44.1.DATASHEET 1 – ICT GUARDIAN ALERT LOG....................................................................................................................211

44.2.DATASHEET 3 – COMMUNITY DRILLS EVALUATION FORM......................................................................................................216
44.3.DATASHEET 4 – COUNTING PEOPLE AT STRATEGIC POINTS....................................................................................................221
44.4.DATASHEET 5 – ICT GUARDIAN DAILY TEST LOG.............................................................................................................223
44.5.DATASHEET 6 – ICT GUARDIAN BREAKDOWN LOG...........................................................................................................225
44.6.DATASHEET 7– EVENT OF INTEREST FORM.........................................................................................................................228
44.7.DATASHEET 8 – HIH MONITOR DRILL PERFORMANCE EVALUATION.......................................................................................229
LIST OF TABLES
TABLE 1 DISCUSSION OF PROJECT ACTIVITIES AND PROGRESS....................................32
TABLE 2 LM-HWS ACCESS AND MESSAGE MANAGEMENT ATTRIBUTES.....................39
TABLE 3 LIST OF ACRONYMS AND ABBREVIATIONS USED IN THIS REPORT..............59
TABLE 4 VILLAGE CATEGORIZATION RELATIVE TO THE TRAINING REGIME AND

ORGANIZATIONAL RANKING.......................................................................................................66
TABLE 5 THE DEPLOYMENT OF ICT COMBINATIONS IN THE 32 SELECTED

SARVODAYA COMMUNITIES.........................................................................................................66
TABLE 6 SYNOPSIS OF SARVODAYA DEVELOPMENT STAGES..........................................67
TABLE 7 CLIQUES AND THE SET OF CORRESPONDING PARAMETERS..........................73
TABLE 8 MATRIX TO DETERMINE MESSAGE PRIORITY WITH CAP ELEMENTS........74
TABLE 9 SCALING FUNCTION FOR FULL-CAP CAPABILITIES...........................................75
TABLE 10 PERCENTAGE OF POPULATION BY ETHNIC GROUP (RACE) LITERACY IN

THE SECOND LANGUAGES.............................................................................................................76
TABLE 11 SCALING FUNCTION FOR LANGUAGE DIVERSITY............................................76
TABLE 12 SCALING FUNCTION FOR AUDIO AND TEXT COMMUNICATION MEDIUMS

.................................................................................................................................................................77
TABLE 13 SCALING FUNCTION FOR ACKNOWLEDGEMENT OF MESSAGE RECEIPT78
TABLE 14 SCALING FUNCTION FOR BI-DIRECTIONALITY.................................................80
TABLE 15 PARAMETERS ASSOCIATED WITH CALCULATING TOC..................................81
TABLE 16 CAP VALUES FOR AN URGENT PRIORITY MESSAGE......................................109
TABLE 17 CAP VALUES FOR A HIGH PRIORITY MESSAGE...............................................109
TABLE 18 CAP VALUES FOR A LOW PRIORITY MESSAGE................................................110
TABLE 19 CAP <ALERT> SUB-ELEMENTS................................................................................111
TABLE 20 CAP <INFO> SUB-ELEMENTS...................................................................................113
TABLE 21 CAP <RESOURCE> SUB-ELEMENT.........................................................................113
8
TABLE 22 CAP <AREA> SUB-ELEMENT....................................................................................113
TABLE 23 CAP MESSAGE <STATUS> VALUES........................................................................114
TABLE 24 CAP MESSAGE <TYPE> VALUES.............................................................................115
TABLE 25 CAP MESSAGE <SCOPE> VALUES...........................................................................115
TABLE 26 CAP MESSAGE <CATEGORY> VALUES.................................................................116
TABLE 27 CAP MESSAGE <EVENT> VALUES..........................................................................116
TABLE 28 MESSAGE <AREADESC> VALUES...........................................................................119
TABLE 29 CAP STATUS FOR TEST CATEGORIES..................................................................122
TABLE 30 CONTACT DATABASE FOR CONFIRMING EVENTS..........................................124
TABLE 31 EVENT AND SITE PARAMETERS OF SELECTED HAZARDS............................132
TABLE 32 ELEMENTS AT RISK....................................................................................................140
TABLE 33 PARAMETERS FOR QUANTIFYING RISK.............................................................140
TABLE 34 GUIDELINES TO DO A MATRIX ANALYSIS..........................................................143
TABLE 35 RISK CATEGORIZATION BY CLASS.......................................................................143
TABLE 36 PARTICIPATORY RURAL APPRAISAL...................................................................152
TABLE 37 TOOLS OF PRA..............................................................................................................152
TABLE 38 STEPS IN PARTICIPATORY MONITORING AND EVALUATION (PME)........154
TABLE 39 DISASTERS ACCORDING TO TIMING....................................................................163
TABLE 40 DISASTERS ACCORDING TO PREDICTABILITY.................................................163
TABLE 41 DISASTERS ACCORDING TO RESPONSE TIME...................................................163
TABLE 42 EVACUATION PREPAREDNESS RESPONSIBILITY............................................165
TABLE 43 ALLOCATION OF ICTS TO COMMUNITIES WITH TRAINING AND

ORGANIZATIONAL CAPACITY...................................................................................................178
TABLE 44 LIVE-EXERCISE DATA FOR FUNCTION: DOWNLOAD,

ACKNOWLEDGMENT, & EOI COMPLETION..........................................................................179
TABLE 45 LIVE-EXERCISE DATA FOR FUNCTION: APPROVAL ISSUE AND OVERALL
9
EFFICIENCY......................................................................................................................................181
TABLE 46 EFFICIENCY DATA FOR HIH MONITOR AND ICT TERMINAL DEVICES PRE
COMMUNITY.....................................................................................................................................182
TABLE 47 EFFICIENCY DATA FOR ICT GUARDIAN AND ERP COORDINATOR IN

COMMUNITY.....................................................................................................................................183
TABLE 48 - EFFICIENCY DATA SUMMARY FOR ALL FUNCTIONS IN TABLES 20 & 21

...............................................................................................................................................................184
TABLE 49 SIGNAL STRENGTH MEASURED IN TERMS OF ENUMERATING BARS FOR

EACH DEPLOYMENT......................................................................................................................187
TABLE 50 MEAN CERTAINTY FOR EACH OF THE ICT DEPLOYMENTS........................189
TABLE 51 DESCRIPTION OF THE ICT CAPABILITIES FOR THE FULL CAP

COMPLETENESS SET OF PARAMETERS..................................................................................190
TABLE 52 RATING FOR THE ICTS FOR EFFECTIVENESS OF RELAYING A COMPLETE

FULL CAP MESSAGE.......................................................................................................................190
TABLE 53 DESCRIPTION OF THE ICT CAPABILITIES FOR THE ACTIVE ALERTING

SET OF PARAMETERS....................................................................................................................191
TABLE 54 RATING OF ICTS FOR EFFECTIVENESS OF ALERTING THE LAST-MILE

COMMUNITIES.................................................................................................................................191
TABLE 55 COSTS ASSOCIATED WITH EACH OF THE ICTS................................................191
TABLE 56 TOTAL COST OF OWNERSHIP PER HOUSE HOLD OVER THREE YEARS. .192
TABLE 57 DESCRIPTION OF THE ICT INTEGRATION IN TO DAILY LIFE OF

COMMUNITY’S PERCEPTIONS....................................................................................................192
TABLE 58 RATING OF ICTS FOR EFFECTIVENESS OF ADOPTABILITY.........................193
TABLE 59 WEIGHT OF THE ICTS IN KG...................................................................................193
TABLE 60 VOLUME OF THE ICTS IN CUBIC METERS..........................................................193
TABLE 61 LIFETIME OF POWER OF ICTS IN HOURS...........................................................194
TABLE 62 RATING OF ICTS FOR EFFECTIVENESS OF MINIATURIZATION.................194
TABLE 63 BI-DIRECTIONALITY OF ICTS IN TERMS OF UPSTREAM AND
10
DOWNSTREAM.................................................................................................................................194
TABLE 64 RATING OF ICTS FOR EFFECTIVENESS OF TWO-WAY COMMUNICATION

...............................................................................................................................................................194
TABLE 65 MESSAGES RECEIVED BY THE RAD AND MOP..................................................199
TABLE 66 SILENT TEST RESULTS 06 TO 10 AUGUST 2007...................................................202
TABLE 67 SILENT TEST RESULTS 15 TO 20 OCTOBER 2007................................................204
TABLE 68 SILENT TEST RESULTS 26 TO 30 DECEMBER 2007.............................................206
TABLE 69 EMAIL FROM EXECUTIVE DIRECTOR OF SARVODAYA ON TSUNAMI

WATCH 12-SEP-2007.........................................................................................................................209
TABLE 70 PERFORMS OF MOBILE PHONES ON THE REAL EVENT OF A TSUNAMI

WATCH ON 12-SEP-2007..................................................................................................................209
LIST OF FIGURES
FIGURE 1 PERFORMANCE SUMMARY OF THE NATIONAL AND LAST-MILE

COMPONENTS....................................................................................................................................17
FIGURE 2 PERFORMANCE SUMMARY OF THE ICTS DEPLOYED IN THE LAST-MILE19
FIGURE 3 HIH FUNCTIONS AND RESPECTIVE AVERAGE EFFICIENCY SCORE...........23
FIGURE 4 EFFICIENCY OF COMMUNITY FIRST-RESPONDERS WITH THE PRESENCE

AND ABSENCE OF ERPS...................................................................................................................24
FIGURE 5 PERFORMANCE SUMMARY OF ICT-G WITH RESPECT TO THE VARIOUS

ICT DEPLOYMENTS..........................................................................................................................25
FIGURE 6 ICT GUARDIANS WITH RESPECT TO COMMUNITY ORGANIZATION

LEVEL AND ICT DEPLOYMENTS..................................................................................................27
FIGURE 7 ERP COORDINATOR EFFICIENCIES WITH RESPECT TO COMMUNITY

ORGANIZATION AND ICT DEPLOYMENTS...............................................................................28
FIGURE 8 EFFICIENCY OF COMMUNITIES GROUPED BY DISTRICT...............................31
11
FIGURE 9 REMOTE ALARM DEVICE...........................................................................................37
FIGURE 10 NOKIA6600 MOBILE PHONE.....................................................................................37
FIGURE 11 AREA RECEIVERS........................................................................................................38
FIGURE 12 – VSAT SYSTEM............................................................................................................38
FIGURE 13 - CDMA FIXED PHONE................................................................................................38
FIGURE 14 DEWNS BROWSER SW................................................................................................41
FIGURE 15 ANNY BROWSER SW...................................................................................................41
FIGURE 16 IPAS BROWSER SW......................................................................................................41
FIGURE 17 SYSTEM ARCHITECTURE OF AN END-TO-END HAZARD DETECTION AND

WARNING CHAIN OF SYSTEMS....................................................................................................61
FIGURE 18 COMMUNITY-BASED ALERT AND NOTIFICATION COMMUNICATION

ARCHITECTURE................................................................................................................................61
FIGURE 19 MICROWAVE INTERNET LINK................................................................................62
FIGURE 20 VSAT INTERNET LINK................................................................................................62
FIGURE 21 ICT PROVIDERS, GATEWAYS, AND TECHNOLOGIES LINKING HIH WITH

THE LAST-MILE.................................................................................................................................65
FIGURE 22 SEQUENCE DIAGRAM OF THE ACTORS AND ASSOCIATED FUNCTIONS

OF LM-HWS.........................................................................................................................................69
FIGURE 23 GRAPH FOR CALCULATING THE EFFECTIVENESS BASED ON SIGNAL

STRENGTH...........................................................................................................................................71
FIGURE 24 EXAMPLE OF SIGNAL STRENGTH BARS..............................................................72
FIGURE 25 ENUMERATION FUNCTION FOR WAKE UP FUNCTION...................................79
FIGURE 26 GRAPHS OF TCO PENALTY FUNCTION................................................................82
FIGURE 27 GRAPH OF THE EQUATION (11) FOR WEIGHTS ABOVE 20KG......................83
FIGURE 28 GRAPH OF EQUATION (12) FOR VOLUME ABOVE 0.125M3.............................84
FIGURE 29 GRAPH OF EQUATION (13) FROM 0 TO 16 HOURS.............................................85
FIGURE 30 COMPONENTS OF THE AREA TERMINAL DEVICE...........................................87
12
FIGURE 31 STATE TRANSITION DIAGRAM FOR THE AREA SYSTEMS............................87
FIGURE 32 HIGH LEVEL ARCHITECTURE OF THE DEWNS WITH NETWORK AND

TERMINAL DEVICES........................................................................................................................91
FIGURE 33 CIRCUIT LEVEL SYSTEM ARCHITECTURE OF THE RAD...............................92
FIGURE 34 EXPECTED OPERATIONAL STATES OF THE RAD & MOP..............................94
FIGURE 35 REAR OF BUC ALB AND FRONT/REAR OF IDERCT MODEM.........................95
FIGURE 36 OPERATIONAL STATES OF THE IPAS ON A PERSONAL COMPUTER (PC).

.................................................................................................................................................................96
FIGURE 37 OPERATIONAL STATES OF 1XRTT CDMA 2000 UNIT.......................................98
FIGURE 38 THE LM-HWS DEPICTED AS A TWO-STAGE RELAY NETWORK..................99
FIGURE 39 ORGANIZATIONAL STRUCTURE OF THE HIH.................................................102
FIGURE 40 SIMPLIFIED PROCEDURE FOR ISSUING WARNING MESSAGES FROM THE

HIH.......................................................................................................................................................103
FIGURE 41 CAP MESSAGE STRUCTURE FOR LM-HWS........................................................111
FIGURE 42 SAMPLE SCRIPT BASED ON CAP ELEMENTS...................................................118
FIGURE 43 DIAGRAM OF RISK MATRIX ANALYSIS.............................................................142
FIGURE 44 DIAGRAM OF THE DISASTER MANAGEMENT CYCLE..................................146
FIGURE 45 LOCATIONS OF 32 HAZINFO PROJECT PARTICIPATING COMMUNITIES.

...............................................................................................................................................................177
13
1. SYNTHESIS
An analysis of the Asian Tsunami of December 2004 showed that thousands of lives could have been
saved if Sri Lanka and other affected countries had effective warning systems in place at national and
local levels. In December 2005, LIRNEasia, with funding support from the International Development
Research Center (IDRC), initiated a research project to evaluate the “last-of-the-mile” communication
component of an all-hazards warning system (HazInfo) for Sri Lanka.
For purposes of the HazInfo project, the research focused on the non-government organization (NGO)
contribution and was designed around a governance structure whereby the non-profit NGO, Sarvodaya,
provided oversight, training, and a hazard information hub for the monitoring of hazard threats and
dissemination of alert messages to local communities within the Sarvodaya network of communities
utilizing combinations of different Information Communication Technologies (ICTs). The Pilot
established Last-Mile networking capability for 32 tsunami-affected communities with 5 ICTs:
Addressable Satellite Radios for Emergency Alerting (AREA), Remote Alarm Devices (RAD), Mobile
Phones (MOP), Fixed Phones (FXP) and Very Small Aperture Terminals (VSAT).
Results to date suggest that the basic internetworking arrangement at lower technical layers has proven
to be reasonably robust and reliable but that a key challenge remains in the upper layers of application
software and content provision. This is evident in the apparent difficulties faced when implementing
Common Alerting Protocol (CAP) messaging over multiple last-mile systems that include commercial
satellite and terrestrial network technologies in modes of voice and text. Lessons learned from Silent-
Tests and Live-Exercises point to several key bottlenecks in the system where the integrity of CAP
messages is compromised due to problems associated with software interoperability or direct human
intervention. The wider implication of this finding is that content standards by themselves are not
sufficient to support appropriate and timely emergency response activities. Those working with content
standards for hazard information systems must consider closely the interoperability issues at various
layers of interconnectivity.
Community-based Last-Mile Hazard Warning System (LM-HWS) Hazard Information Hub (HIH) and
Community First-Responders are expected to have Reliability above 90 percent, giving the
Communities adequate time to execute their Emergency Response Plans, and not at 77 percent as the
system is performing now. This aspect implies that the first responders require rigorous training and
certification in emergency communication to avoid ambiguity and misinformation, which is a
necessary condition to supplement the deficit of an end-to-end automated alerting system. Further
findings indicate that; reliability at the central message relay can be enhanced by introducing a
Multilanguage single input multiple output software application (i.e. P2P Multilanguage CAP Broker).
The ICTs as stand alone communication systems have 97 percent reliability but have merely 78
percent reliability as warning technologies. Based on the project defined indicators to rate the
effectiveness of the ICTs, they were assessed to average 36 percent effective as warning technologies.
However, overall performance of the alert and notification last-mile terminal devices is enhanced when
a community is equipped with a technology combination that enhances “complimentary redundancy”
in reliability and effectiveness; where the combination of the WorldSpace AREA sets and the Dialog/
Microimage Java-enabled MOP performed the best with the AREA + FXP following closely.
14
2. RESEARCH PROBLEM
2.1 Project Objective
“It is easier to be wise with the benefit of hindsight. Yet, lessons learnt -- negative and positive – help
in the better design and implementation of future projects. The HazInfo project, which was an action
research effort meant to learning while doing, brings out a few important lessons that are summarized
in this report1.”
The overall aim of the project was to set the stage for community-driven initiatives at the last mile of
the hazard information dissemination system (HazInfo). The proposal states that it is not an
implementation project but a pilot study, based on experimental research design. The function of the
pilot study was partly a technology assessment to determine the most cost effective and reliable
solutions for a LM-HWS, including an evaluation of their integration into the everyday activities of the
community2.
An aspect of the project was to expand upon the results of the Last Mile Hazard Warning System by
engaging disaster risk reduction techniques in Sarvodaya villages to augment village resiliency and
responsiveness to disaster while integrating community-based disaster management through the
Sarvodaya Community Disaster Management Center (SCDMC).
2.2 Project Background
The research was based on a “participatory concept paper” (NEWS:SL) developed in the months
following the 2004 tsunami. The paper notes that a national early warning system (NEWS) is a “pure
public good.” As such, responsibility for its supply would normally fall on the government. However,
due to lack of capacity for the foreseeable future, it is unlikely that the last mile of such a system will
be provided by the local government or by private firms operating in the marketplace; see Figure 17 in
Technical Annex 2 for an illustration of the last-mile system relative to the end-to-end warning system.
The project was, therefore, designed around a governance structure whereby a non-profit, non-
governmental organization—Sarvodaya—provided oversight, training, and an information hub for the
monitoring of hazard threat conditions and for the initiation of warning messages to local communities
in the Sarvodaya network of villages. Designated first responders selected from the local communities
were responsible for overseeing emergency preparedness, message dissemination, and emergency
response at the local level. Figure 18 in Technical Annex 2 illustrates the Sarvodaya community-based
warning system.
An important aim for the study was to address the rural-urban distinction associated with the
penetration of communications technology in the Eastern Province and Western Province of Sri Lanka.
Disparity in access to communications media in the rural-urban divide is a significant obstacle to
1
This a quote directly extracted from the report presented to the HazInfo Project by the Training Partner: TVE Asia Pacific (http://www.tveap.org)
2
Gordon Gow, PhD, University of Alberta, Canada, ggow@ualberta.ca ,
15
achieving effective warning in the last mile of the system in Sri Lanka. Access to resources and
training is also split along this rural-urban divide.
2.3 Comments on the Proposed Research
The primary objective was to evaluate the suitability of various ICTs as the basis of a LM-HWS in Sri
Lanka. As illustrated in Tables 4 and 5 in Technical Annex 4, an experimental design was proposed to
assess the technology. Six factors were to be considered:
• Reliability of the ICTs

• Effectiveness of the ICTs
• Effectiveness of the training regime
• Level of organizational development
• Gender specific response
• Integration of ICTs into everyday life
These factors were assigned a set of corresponding indicators (see Technical Annex 5 to 11) that
formed the basis for observations and evaluations of the technology and training. It is important to
note that the combination of technology and training involved the specification of distinct but
possibility related criteria. In some cases, technology was the critical factor, whereas, in other cases,
training was most significant. In most cases, however, some combination of technology and training
was responsible for observed results.
Simulated drills took place over 6 month period to gather information with respect to the six stated
specific objectives above and following hypotheses:
• Stage 4 & 5 Sarvodaya villages that are more organized, i.e., have a formal structure that enables
coordination and direction of activities will respond more effectively to hazard warnings than less
organized stage 1, 2 & 3 villages.
• Villages that are provided training in recognizing and responding to hazards along with
deployment of ICTs will respond more effectively to hazard warnings than villages that received
no training.
• Villages that have ICTs deployed for dissemination of hazard information will respond more
effectively to hazard warnings than villages that have to rely on their existing channels of
information for warnings.
• ICTs that in addition to their hazard function can also be leveraged in other areas to enrich the lives
of the villages will potentially have lower downtime than ICTs that are poorly integrated into the
day to day life of the beneficiaries.
Of the initially chosen 10 districts, Colombo, Kalutara, Galle, Matara, Hambantota, Ampara,
Kalmunai, and Trincomalee districts successfully completed their Live-Exercises. However, as a result
of the North-East conflict, inadequate leadership, lack of resources, and language barriers prevented
Jaffna and Batticaloa District from successfully completing training, deploying ICTs, and conducting
Live-Exercise. Of the fully participating communities, all had an equal level of participation and
16
enthusiasm implying that Sarvodaya villages in all stages responded at the same capacity to hazard
warnings and disaster resilience capacity building; i.e. all executed tsunami evacuation plans, which is
the only emergency response plan the community was trained to execute.
Last-Mile Communities realize the potential of the LM-HWS to help them sleep easy at night.
Revoking it without enhancing it based on the research findings and also expanding into the remaining
15,000+ Sarvodaya Communities in Sri Lanka would jeopardize the trust they have in Sarvodaya.
3. RESEARCH FINDING
Indicators of the overall ICT Performance were assessed against the point system based on the
composition of a set of scaling functions as described in the Technical Annex 5 – 13 and the expected
outcome of the ICTs are discussed in Technical Annex 14. Reader is encouraged to read the Technical
Annex prior to continuing this section to better understand the assessment methodology that gives rise
to the results discussed in this section. The results are based on a very small sample space and do not
show the best comparatives. However, the rating provides the planners with a set of indicators to
enhance the reliability and effectiveness of individual components: HIH Message Relay, ICT
Networks/Terminal-devices, and Community, of the National and Last-Mile systems; where Figure 1
illustrates a summary of the reliability and effectiveness of the mentioned three components.
Average Reliability and Effectiveness of Independent

Components
0.83
0.76 0.78
0.64
0.36
0.10
HIH ICT Community

Component Name
Reliabiity Effectiveness
Figure 1 Performance summary of the National and Last-Mile components
The overall performance of the “HIH” in Figure 1, synonymous with the National Warning Center, is
acceptable. However, additional training and drills must be conducted for the staff to be well versed
with the procedures and to be able to carry out the tasks within the benchmark time. Section 3.2
17
discusses gaps in the HIH component. “Pink” components in Figure 17 & 18 contribute the HIH.
Effectiveness of the “ICT” component in Figure 1 was well below the expected requirements to be
used as an early warning technology. Although the reliability of the ICT component is relatively high it
does have several gaps that need to be filled in order to ensure communication certainty and efficiency.
Section 3.1 discusses the assessment and analysis of the ICT component in terms of reliability and
effectiveness. “Yellow” components in Figure 17 & 18 are what contribute the ICT segment.
“Community” segment, in Figure 1, needs to be focused on the most. The live-exercises revealed that
inadequate training and improper notification resulted in unstable behavior. Sections 3.2, 3.3, and 3.4
discuss the performance of the Community component. “Blue” components in Figure 17 & 18 are what
contribute to the Community segment.
With the LM-HWS taking the approach of all-hazards all-media alerting and notification, it is vital that
unambiguous full CAP messages are issued by the HIH and are received by the Communities. As a
result it is recommended that policies are implemented to provide formal training and certification of
the HIH-M and ICT-G to be absolutely competent of communicating with accuracy and follow proper
procedures. Further exercises must be conducted to realize the true potential of utilizing a Community-
based system; especially for the Last-Mile. Since the data reported is from the first trial of its kind it is
impossible to adapt methodologies such “Markov Decision Processes” to improve the policies because
of the insufficiency of probabilistic data to develop transition probability tables to determine the
effectiveness of the policies in practice. However, through conventional trial and error methodologies
the pilot project is able to suggest intuitive policies that can strengthen the performance of a National
Early Warning System (NEWS).
3.1. Performance of the ICTs as a Warning Technology
This section combines the analysis of the specific objectives related to the technology: reliability of the
ICT as a warning technology and effectiveness of the technology as a warning technology.
Methodology applied to calculate the reliability of the technology in terms of efficiency and certainty
is described in Technical Annex 7. The 11 indicators used to evaluate the effectiveness of the ICTs are
discussed in Technical Annex 8.
There is a significant gap between the ICT Reliability as stand alone ICTs and as warning technologies
in last-mile communities. The “Reliability of ICT”, in Table 47 in Appendix D, presents the findings
as to whether the technology worked on the day of the drill. The average Reliability of ICT in the Last-
Mile, recorded in Table 47 (ICT Terminal Devices) in Appendix D, reveals as to how the communities
with the particular ICT performed overall during the drills. The number of ICTs distributed in the
research matrix (see Tables 44 in Appendix B) is not uniform and also the number of trials conducted
for each ICT is not equal.
Despite the envisaged light-speed transmissions in the ICTs, a standard message transmission took
approximately 7 seconds to carry a text message from the HIH to the Community, there were delays
caused mainly by human errors such as the First-Responder message recipients in the community,
namely the ICT Guardians (ICT-G) not correctly aligning the antennas of the satellite system,
18
accidentally deleting the Java applet on the mobile phone, wrongly configuring the application that
restricted receipt of certain levels of hazardous events, ICT-G not being close to the wireless terminal
devices to receive the message on time, or simply the ICT Provider terminating the services for not
having received a payment for the monthly bill. Tables 67 to 69 in Appendix H show the issues for
ICTs failing during Silent-Tests due to the same reasons of ICT-G unavailability, disruption to
services, or Terminal Device not always-on and ready to receive alerts. An unusual occasion was the
MOP and FXP failing on the day of the exercise as a result of the conflict situation in the North and
East of the Island; where the Government Military had instructed all commercial wireless operators,
GSM and CDMA, to shut off cells in the respective battle zones. However, the unidirectional AREA
satellite based sets functioned in the war-zones when the terrestrial technologies were shut off. As
mentioned previously communications in English in the rural communities were difficult for the locals
to comprehend. The Active Alert function was not up to par to grab the attention of the ICT-G.
Restrictions of the terminal capabilities of carrying a full message caused false information
propagation in the communities.
Figure 2 Performance summary of the ICTs deployed in the last-mile
Comparison of Reliability and Effectiveness of ICT as a

Warning Technology in a LM-HWS
$0.75
AREA+MOP 0.71
$0.75
AREA+FXP 0.89
$0.43
AREA+RAD 0.71
ICT Configuration
$0.05
AREA 0.59
$0.24
MOP 0.27
$0.26
FXP 0.47
$0.09
RAD
$0.04 Effectiveness
VSAT
Reliability
Control Group 0.15
The efficiency outcome of the individual ICTs during live-exercises is given in Table 47, Appendix D.
Certainty measures of the individual ICTs recorded on the day of the exercise (live-exercises) at each
location are stated in Table 50, Appendix E. The data pertaining to the effectiveness of the ICT
Terminal Devices with respect to the 11 indicators is recorded and discussed in Appendix F. Figure 2
summarizes the reliability and effectiveness of the ICT deployments in the selected communities.
Figure 2 shows when individual ICTs: AREA, MOP, FXP, RAD, and VSAT are deployed in the
19
communities the reliability and effectiveness are quite low. However, when they are coupled such as
the AREA + MOP and AREA + FXP then the performance is relatively high. The AREA is quite
ineffective as a result of the one-way capability but is relatively very reliable as a satellite technology.
The MOP is relatively effective since it is capable of displaying alert text messages in Sinhala, Tamil,
and English; however, the MOP, as a stand alone, is an unreliable terrestrial technology. When both
the AREA and MOP are coupled they complement each other in terms of their respective strengths in
reliability and effectiveness, which is termed as “complementary redundancy”.
The reliability scores for the individual RAD was not available because according to ICT deployment
in the communities, Table 44 in Appendix B, a stand alone RAD was not deployed in a community. In
the case of the VSAT the terminals were not deployed for the community to use the ICT during the
live-exercises. The Control Group was not provided with an ICT by the project. However, the
communities used available ICTs such a MOPs and FXPs to establish communication with a
neighboring community in their district that participated in the live-exercises.
Hypothesis: Villages that have ICTs deployed for dissemination of hazard information will respond
more effectively to hazard warnings than villages that have to rely on their existing channels of
information for warnings.
Conclusion: All 4 Control Villages that took part in the Live-Exercises had made an alliance with
the neighboring Sarvodaya community to receive a telephone call by the Community Chairperson
(potential ICT Guardian). The messages were received over their personal GSM mobile or wireless
CDMA fixed telephones in the form of a voice call. The Control Villages had prepared in advance to
receive the messages. The Reliability of the 4 Control Villages lag because they did not receive the
alert directly from the HIH. It is not sure whether the Control Villages would perform this well if it
was not a drill or was an unplanned random event. The Reliability of both Communities with ICTs
and without ICTs was too weak to make a clean comparison between the 2 sets and support the
hypothesis.
3.2. Contribution of the Training Regime
Training Regime in HIH
The training and operational procedures for the HIH is discussed in Technical Annex 15 – “Guidelines
for the HIH”. As discussed in Technical Annex 2 and 15, the HIH – Hazard Information Hub (pink
block in Figure 18) operational procideures are similar to that of a National Early Warning Center
(pink block of Figure 17) The set of HIH functions are -- downloading bulletines from gloabal/local
feeds, Acknowledging reciept, generating and EOI (see Datasheet 7 in Appendix I), getting pre-
dissemination approvals, and issuing CAP message. With some procedural improvement and training
the HIH functions, up to the point of relaying (issuing) the CAP Message, can be made to perform over
0.90 efficiency. Howevere, the bottleneck is in the final step of issuing the message via the ICT
interfaces. Some of the delays were caused by login failures as a result of the HIH staff not
remembering the multiple passwords or the interface access password had expired and needed
20
administrative assistance to reset it. As illustrated in Figure 3, the score of 0.69 for the Relay function
is mainly a result of having to operate a multiple of Single Input Single Output (SISO) message relay
software interfaces, which is intuitively a cumbersome task. Data shows that the efficiency drops
exponentially when the number of independent applications the HIH has to use increase as a result of
repetition of entries. Interfaces that require recording voice messages are a main bottleneck that brings
down the efficiencies due to the time consuming recording and editing processes.
Average Reliability of HIH Monitor's Functions
1.00 0.97
0.86
Reliability
0.69
0.41
t
en
d
ay
I
va
O
oa
em
el
ro
nl
R
pp
dg
ow
A
le
D
w
no
ck
A
Name of Function
Figure 3 HIH functions and respective average efficiency score
Applying the methodology in Technical Annex 5 for measuring reliability as a function of efficiency
Tables 45 & 46 in Appendix C provide the set of data from which the HIH average reliability is
calculated. Figure 3 shows the average efficiency for each HIH-M function’s score obtained from all
the live-exercises; where the Acknowledgement and Relay functions are poorest. The low score of 0.40
on acknowledgement is mainly due to over excitement during the exercises and the HIH staff
forgetting to follow standard procedures. Therefore, such important tasks must be made mandatory
through technology by forcing the staff members to complete the task, else disabling the consequent
tasks.
Emergency communication text messaging, in the live-exercises, was mainly in English, except for the
Java enabled Multilanguage mobile phone, which carried Sinhala, Tamil, and English. Therefore, the
complexity of translating the English CAP messages to Sinhala and Tamil is not reflected in the data
too well. However, it is intuitive that without any intelligent real-time natural language translator it
would be very difficult for the HIH-M (HIH staff) to achieve expected the 90% Reliability benchmark;
especially for all-media all-hazards communications in multiple languages.
EXAMPLE 1: Assuming the duration of a hazard such as a tsunami initiated in the Indian Ocean is 90
21
minutes, from the time of detection to the time of impact, by applying the methodology (i.e. equation
(3)) for calculating efficiencies, described in Technical Annex 5, an efficiency of 0.78, shown in
Figure 1 translates to the HIH taking 21.6 minutes, which is too long and takes away valuable time
from the community to activate their evacuation plans. An efficiency of 0.90 would translate to 9.0
minutes, which is what is expected from the HIH. In terms of effectiveness the scoring had dropped
below the 0.90 benchmark because the HIH did not populate essential attributes of the EOI form
properly. Methodology for evaluating the HIH-M effectiveness is described in Technical Annex 6.
There were also inconstancies in competency levels among different HIH staff scheduled to participate
in the respective live-exercises.
Training Regime in Community
There are two sets of Community-First-Responders; namely the ICT-G and the ERP-C. The training
content for the ICT-G and ERP-C is taken from Technical Annex 15 and Technical Annex 16 – 23,
respectively. The ICT-G simply relay the messages to the ERP-Cs who intern relay the message to the
community households. ICT-G have a point-to-point communication protocol; where as the ERP-C
must communicate the message to multiple points (i.e. all the households). Therefore, the reliability
results, in Table 48, for the ERP-C compared to ICT-G are significantly low because the ERP-C did
not broadcast the message but distributed the message from house-to-house either by word-of-mouth
(diffusion effect) or with a mobile Pubic Announcing (PA) system. In some cases they used
Temple/Church bells and Mosque PA system to broadcast the message. However, these schemes were
ineffective because the households did not hear the Temple/Church bell and the ERP-C had to repeat
the dissemination process by sending ERP-C out to inform the households by word-of-mouth.
Efficiency of First-Responders in Communities

w.r.t Emergency Response Plans
0.56
0.50
0.46 0.43
Efficiency
With ERP
No ERP
ICT-G ERP-C
First-Reponders
Figure 4 Efficiency of Community First-Responders with the presence and absence of ERPs
Once again, evaluation of the Community segment for reliability as a function of efficiency was
calculated using the methodology discusses in Technical Annex 5. The effectiveness of the ICT-G and
the ERP Coordinators were measured based on the methodology discussed in Technical Annex 6.
22
Figure 4 shows that the average efficiency scores of the community First-Responders between
communities that received formal ERP training versus communities that did not receive ERP training
does not show a significant disparity. The data for this query is obtained by relating Table 44 in
Appendix B, of communities that had received formal ERP training versus that didn’t, with a
combined ICT-G and ERP-C efficiency data in Table 48 in Appendix D. Figure 4 shows that ERP-Cs
in communities that received formal ERP training rank higher than those communities that did not
receive ERP training. However, the efficiency scores are not close to the expected 0.90 benchmark. All
the ICT-Gs received training on operating and up keeping the ICTs but ERP training was only given to
those who were belonged to the villages that received ERP training. Regardless the performance
expectation levels were very poor. The ICT-G had to be coached during the live-exercises to get them
to perform their tasks properly.
Efficiency of ICT-G w.r.t ICT Deployments
FXP 0.94
MOP 0.94
ICT Deployment
AREA 0.87
AREA+RAD 0.57
AREA+FXP 0.90
AREA+MOP 0.91
Efficie ncy
Figure 5 Performance summary of ICT-G with respect to the various ICT deployments
Figure 5 shows a comparison of the various ICT deployments in the communities and the efficiency of
the ICT-G in using the ICTs. ICT-G in communities with an AREA+RAD show a significant reduction
in the efficiency; where as all other deployments are quite similar and have very little variance in their
efficiency scores. RAD was the only device that was a prototype innovation introduced to the project
and field tested for the first time. It is apparent from Figure 5 that off-the-shelf readly available ICTs
were easily adopted in to the communities; where as the newly introduced RAD was found to be
cumbersome to adopt; as a result dragged the score of the AREA+RAD deployment significantly
down. Stand alone AREA is slightly below the 0.90 expected benchmark because the AREA is the
only device that carried a substantial segment of the CAP text message. Therefore, the ICT-G had to
spend a few more extra minutes recording the information in the Alert Log (see Datasheet 1 in
Appendix J) relative to the other ICT Terminal Devices which did not carry a significant amount of the
CAP message but rather a very short message. Results in Figure 5 imply that the ICTs are
23
unsophisticated and ICT-G can be easily trained to adopt the ICTs as warning device in the
communities.
ICT-Gs were forced to decode partial CAP messages received by ICT terminals that could not carry a
full CAP message. Shortcomings of the ICT terminals with respect to displaying full CAP messages
were a major challenge of Internetworking with CAP for relaying of complete messages. As a result of
the partial messages, in many of the live-exercise, mutation of information was witnessed. When the
HIH had issued a “Category 4 Cyclone” alert the communities executed tsunami evacuation plans; i.e.
running to a higher grounds when they were actually supposed to seek shelter at lower ground.
According to the information in the alert message and the time the ICT-G received the message, the
communities had approximately 4-6 hours to prepare for the arrival of the Cyclone. However, the
communities thought the hazard was a short fuse type hazard (i.e. tsunami) and ordered immediate
evacuation of the community.
The weakness can be blamed on the quality of the training the community had received in developing
their ERPs by the Shanthi Sena HazInfo Trainers. It is incorrect to put the blame on the Communities
because the CAP structure is constructed to lodge all-hazards all-media unambiguous messages; where
the ICT-G is supposed to provide accurate instruction to the ERP Coordinators in order to execute
precise ERPs and prevented the community from executing the wrong ERPs. Except for ICT-Gs who
were equipped exclusively with an AREA, all other ICT-G had bi-directional means to communicate
with the HIH (upstream communication). Hence, when the ICT-Gs were doubtful of the meaning of
the message they had the option of contacting the HIH or an alternate source to verify the alert
communication. In all of the live-exercises none of the ICT-G took the responsibility to either
acknowledge receipt of message or attempt to obtain full or further instructions from the HIH.
Researchers realize that there is a rather significant liability of false information diffusing rapidly in a
community if the problem of information mutation at the HIH, ICT-G, and ERP-C stages is not
corrected. Further acknowledge that a comprehensive study of the problem related to information
mutation and chaotic behavior due to false information diffusion in a LM-HWS must be studied. Such
irregular social phenomena can be studied using techniques such as “patterned chaos forecasting”
techniques.
Hypothesis: Villages that are provided training in recognizing and responding to hazards along with
deployment of ICTs will respond more effectively to hazard warnings than villages that received no
training.
Conclusion: The nature of the Live-Exercises could not determine the effectiveness of the training
regime. Overall, it was observed that the response competency level resulting from training was way
below expected level. The drills carried out in the communities were predominantly staged by the
organizers. Since the organizers were the Shanthi Sena HazInfo Trainers the outcomes of the
simulated exercises were identical in each of the communities and shows nor disparity between the
set of Trained and Untrained Communities. However, the project found that training was imperative
for obtain good quality simulation results, reinforcing community emergency planning and raising
community awareness about hazards and interest in local risk management.
24
3.3. Contribution of Village Organizational Development
ICT Guardian Efficiency w.r.t ICT deployment and

Community Organizational level
AREA+RAD 0.91
0.80
AREA+FXP 0.88
ICT Deployment
0.48
AREA+MOP NA 0.91
NA
FXP 0.94
0.93
MOP 0.94
0.84
AREA 0.89
Efficiency
Stage 2 Stage 4
Figure 6 ICT Guardians with respect to community organization level and ICT deployments.
Originally the proposal had partitioned Sarvodaya Stage 1, 2, & 3 villages as less organized and Stages
4 & 5 as more organized communities; description of the Sarvodaya Stages can be found in Table 6 in
Technical Annex 4. However, there were no Stage 5 communities in Sarvodaya in the HazInfo project
selected coastal districts. Remarkably there were no stage 1 or 3 communities that participated in the
project either. The development stage of each village can be obtained from Table 44 in Appendix B.
Since the community comprises the ICT-G and ERP-C activities, a comparison of the efficiencies of
the two actors are illustrated in Figure 6 and Figure 7.
The label NA – Not Applicable implies that there were no communities with the ICT deployments in
communities within the respective organizational category or no data available for those elements. The
AREA+MOP were deployed in a Stage 2 Community in Jaffna district where live-exercises couldn’t
be carried out due to the North-East conflict. Both the Stage 4 communities Inidivina in Hambantota
district and Periyakalar in Batticaloa district did not have the capacity to organize live-exercises and
participate in the drills. From Figure 6 we may infer that Stage 4 communities are better at adopting a
combination of ICTs such as an AREA+RAD or AREA+FXP than Stage 2 communities; the inference
would have been stronger if the case was also true for AREA+MOP. The lack in Stage 2 communities
adopting the combination of ICTs may be because the AREA and the RAD are a newly introduced ICT
opposed to the readily available ICTs such as the MOP or FXP and the less organized communities do
not have the organizational capacity to adopt new technologies. However, the efficiency levels are
close to benchmark acceptable levels of 0.90. Therefore, it is impossible to draw any conclusions based
25
on the organizational development level of a community and the ability to adopt ICTs for early
warnings.
ERP Coordinator Efficiency w.r.t ICT deployment and Community

Organizational level
None 0.69
0.56
AREA+RAD 0.76
0.67
AREA+FXP 0.76
ICT Deployment
0.58
AREA+MOP 0.64
NA
NA
FXP 0.89
MOP 0.70
0.59
AREA 0.74
0.54
Efficiency
Stage 2 Stage 4
Figure 7 ERP Coordinator efficiencies with respect to community organization and ICT deployments
Form Figure 7 in all cases except for the AREA+MOP and FXP the Stage 4 organized communities
performed better than the less organized stage 2 communities. On the average the variance between the
average efficiencies of the set of stage 4 communities and stage 2 communities is approximately 0.14.
Based on the values from Example 1, a 0.14 efficiency margin implies a 12.6 minute difference.
However, all the communities with the particular ICT deployments were quite close to the benchmark
efficiency of 0.60. Deployments where a none readily available novel ICT was introduced such as the
AREA, AREA+FXP, and AREA+RAD, the efficiency gap between the Stage 2 and Stage 4
communities are larger relative to the common MOP. Once again the inference could have been
stronger if similar results were shown in the case of the FXP. The NA – Not Applicable label in the
case of the FXP and the AREA+MOP are due to unavailability of data. The reasons are similar to the
case of the North East conflict and lack of organizational capacity disused in the Figure 6.
Hypothesis: Stage 4 & 5 Sarvodaya villages that are more organized, i.e., have a formal structure
that enables coordination and direction of activities will respond more effectively to hazard
warnings than less organized stage 1, 2 & 3 villages.
Conclusion: communities with adequate capacity and organizational structure in their respective
Districts proved effective in organizing all project activities as well as adopting new technologies.
However, organized communities responded to warnings in the same capacity as the less-organized
26
communities.
3.4. Gender Specific Response to Hazard Mitigation Action
Data from Live-exercises reveals that over 72% Adult participants were Female because the
simulations were conducted between 9am and 12pm where most Men were occupied with their jobs.
For example, the Fishermen would return home around 11am after selling their catch to the merchants
and cleaning their nets/boat. Therefore, only the women could participate in the activities. The women
showed enthusiasm and willingness to participate in all disaster management activities.
In the rural communities females participated more in community level workshops and training. The
outcome is similar to the split indicated in gender participation accounted during the live-exercises.
Once again the down fall of male participation levels were due to adult males engaging in “bread-
winner” activities.
Participation of female and male in organizational activities in preparing the communities was equally
distributed. The volunteers assisting in the organizational activities at the community level were
Shanthi Sena youth. Membership in the Shanthi Sena community volunteer force male and female
participation is equal.
Given the cultural aspects of females being overlay protected; between males and female Shanthi Sena
HazInfo Trainers; i.e. the community ERP trainers, the project found that male trainers were far more
versatile and effective as they did not face the same cultural barriers as the females. In most occasions
the ERP trainers had to travel outside of their community to far rural places as well as stay overnight,
which is not encouraged by parents of female youth trainers. Male trainers were also more vocal and
aggressive in getting the training completed.
3.5. Degree of Integration of ICTs in the daily life of villages
Mobile communication component has functionality that needs to be exploited by the communities.
The Nokia 6600 Java enabled MOP was used in the project. For instance the “MiDews” applet
developed by Microimage would be installed by downloading the applet via GPRS on to the handset
and installing to receive DEWN alerts. Hence, the ICT-G was given access to the internet via GPRS. A
popular feature was downloading and storing MP3 music files to listen over the MOP. Voice calls was
the dominant feature used; where the ICT-G would spend approximately USD15 per month on the
average.
CDMA2000 1x_RTT sets are versatile with FAX, SMS, and Internet capabilities. The Internet
capabilities have proven to be effective in accessibility, speed, and cost. However, the mobility of the
CDMA2000 sets is restricted during Internet access because the communication requires AC power.
Only one ICT-G in Batticaloa district used the facility to access the Internet. The project sees that the
CDMA2000 sets can be affective in both District Centers and the Community Centers. The Mobile
27
Phones are useful for coordinators at all levels.
The WorldSpace Satellite Radios have been effective in providing audio content related to Sarvodaya
activities directly to the communities. Also the communities have been using the Sarvodaya Talk audio
channel to express their opinions on the ongoing projects in their Communities. Ampara and Batticaloa
Districts witnessed power outages during the Live-Exercise but received the alerts over the AREA,
which was powered by 2 AA Batteries.
The VSAT perhaps is the most highly utilized ICT of the 5 ICTs. It was proven to function on UPS
battery power during power failures. The Sarvodaya Community Disaster Management Centre staff
used the high bandwidth internet link with an internal hard-wired and wireless network for Skype,
email, and Internet services necessary to generate and issue alerts. However, VSAT is too expensive
for a community to operate on their own funds.
Overall every ICT has proven to have a unique feature that contributes to the daily functions of the
community. Basically if ICT is given and the usage cost is very low then they will use it.
Hypothesis: ICTs that in addition to their hazard function can also be leveraged in other areas to
enrich the lives of the villages will potentially have lower downtime than ICTs that are poorly
integrated into the day to day life of the beneficiaries.
Conclusion: The live-exercises were carried out with prior notification. Therefore, the hypothesis
cannot be challenged because the ICT-G had the ICTs powered on. However, the silent-test reveals
that the hypothesis is true because ICTs with downtime did not respond to the silent-tests (see
Appendix H)
3.6. Local barriers in responding to alerts
Although the following assessment was not required the researchers believe that indicating the results
on geographical distribution may be of importance for future planners. The District level summary was
a query from Table 48 in Appendix D taken by averaging the overall efficiency of the communities in
each District. The message this section attempts to exemplify is the importance of relaying precise
hazard information in a timely manner for communities to take proper actions to overcome the barriers
described in the following paragraph.
Batticaloa district does not have data as they did not participate in the Live-exercises. Households in
Communities of Ampara, and Hambantota districts are sparsely populated (i.e. scattered and far apart).
Therefore, the time taken to disseminate the hazard information locally via a Public Announcing (PA)
system and completing evacuation were timely. Communities in the towns of Kalmunai district are
densely populated with houses enclosed by high boundary walls. Most houses are trapped in between
several houses where people have to walk through several back-yards to access the road. Communities
in the districts of Kalutara, Matara, and Colombo had the obstacle of crossing the “Galle-road”, one of
the busiest roads. Also the railroad travels alongside the Galle-road. In most cases the people trapped
between the ocean and the transportation infrastructure did not have a direct path to the inland
28
evacuation points but had to traverse along the coast line to find a gap to cross the roads and railway
tracks to access the evacuation routes. Both Galle and Trincomalee district communities do not face
these same barriers. In Galle district, both the railway track and the main road are closer to the coast
line with the houses situated inland. Communities in Trincomalee district do no face the barriers faced
by the other districts discussed.
Average Reliability of Community

Perfromance by District
0.89
0.77
0.68
Reliability
0.57 0.61
0.54 0.53
0.48
NA
ai
a
a
ee
bo
ra
le
a
lo
ot
ar
ar
un
al
pa
al
m
ca
nt
ut
at
G
m
m
lo
ba
al
M
tti
co
al
A
Co
K
am
Ba
in
Tr
H
Distrcit Name
Figure 8 Efficiency of Communities grouped by District
Communities in Kalutara district used Church bells and Mosque PA systems as a local dissemination
method. Both religious facilities were located along the coastline. The wind prevented the sound from
traveling to the households. People did not hear the local message and delayed in responding.
Communities in Matara district area community households span along the coast line. The message
was delivered to the households via a PA system mounted on a three-wheeler (tuk-tuk). The
households that participated in the live-exercises were stretched along a 10km distance. The three-
wheeler with a PA system was inefficient in disseminating the hazard information at a local level. In
the real event of a rapid onset hazard such a Tsunami the tuk-tuk with PA will be inefficient because of
the impracticality of the logistics involved with setting up such a system. Colombo district
communities have a similar geographical setting and distribution of households as Matara district.
Colombo district practiced the same local dissemination method as Matara communities and
experienced the same inefficiencies.
29
4. FULLFILLMENT OF PROJECT OBJECTIVES
Project Start Date: 01-December-2005

Project End Date: 31-November-2007
Project Progress: 92% Completed
Table 1 Discussion of project activities and progress

Activity Start-Date / Deliverable (Document
(% Completed) Outcome End-Date type and Authors)
Identification of Sarvodaya and LIRNEasia select 32 06-Dec-2005 Village and ICT allocation
villages villages from the 226 tsunami spreadsheet with
affected Sarvodaya villages that 25-Jan-2006 Community Name and
(100%) reflect geographic, ethnic, Community Sarvodaya
socioeconomic diversity and different Chairperson’s contact
levels of infrastructural development. details (Spreadsheet,
Author: Nandana
Jayasinghe, Ravindra
Kandage)
Matching See Appendix for Village ICT 27-Jan-2006 Village ICT allocation
technology to allocation table spreadsheet with
village The ICT Guardians were 03-Feb-2006 Community name and ICT
recommended by the Community Guardian’s Contact details
(100%) Chair and District Coordinator for (Spread Sheet: Jananjaya
that Village De Silva)
Supervision of Dialog Telekom GSM Mobile 03-Apr-2006 Hardware and Software

LM-HWS Communication Research Lab will Inventory
deployment by build, test, and deliver 7 units. Since 10-Aug-2006 (Spreadsheet, Author:
external vendors the RADs were in their prototype Jananjaya De Silva)
stage there were many breakdowns
(95%) and the equipment had to be sent
back to the lab for repair. (100%)
30

There were no problems with the Administration, Training,
Java enabled Mobile Phones. There and User Manual
were difficulties in remotely (Text Document, Authors:
installing the Java Alerting Applet. S. Rangarajan,
Microimage-Dialog provides a link to Shanmugaraja, Dileeka
access, download, and install the Dias, P. Warnakulasuriya,
applet through the Internet. However, Nabil Seddhig)
the poor strength of the GPRS signal
made it difficult for the ICT
Guardians to install the applet in the
remote villages. Therefore, they had
to be recalled to Colombo and the
applet was installed for them at the
HIH. (100%)
CDMA Wireless Fixed Phones with Training workshop for the

1x_RTT capabilities were provided. District Coordinators and
However, the SMS, FAX, and Community ICT
Internet capabilities were not enabled Guardians
for the communities. (100%) (Oral and Visual
Presentation, Trainers: S.
All WorldSpace AREA-B systems Rangarajan, S.
were received were given to the Shanmugaraja, Gamini
Community ICT Guardians and Dis- Jayasinghe, Mishan
trict Coordinators after testing them Warnakulasuriya, Nuwan
all at the HIH. (100%) Waidyanatha)
The Main VSAT Station was in-

stalled by Innovative Technologies
and commissioned in October 2006.
However, the Community VSAT Sta-
tion was not installed due to major
delays by the vendor. Therefore,
Live-exercises could not be carried
out in the Community intended to
have a VSAT station. (65%)
31

Creating a hazard The Hazard Information Hub (HIH) 19-Dec-2005 DM Institute Infrastructure
information hub was setup with office furniture, Inventory (spread sheet,
workstations, conference room, and 25-Apr-2006 Ravindra Ariyawickrama)
(100%) other peripherals such as photocopy,
digital camera. Staff was hired and Hardware and Software
trained. Building infrastructure such Inventory
as electricity, water-sanitation, and (spreadsheet, Niluka
communication was setup. Wickramasinghe,
Jananjaya De Silva)
Human Resource Roles

and Responsibilities
(Text Document, Author:
Nandana Jayasinghe)
HIH Performance
Assessment Report (Text
Document, Author: Peter
Anderson)
Training of Community Resources and District 03-Apr-2006 Guidelines for HIH and
villagers to Resources were trained at Sarvodaya Operational zing LM-
maintain LM- Head Office, at District level, and at 11-Sep-2006 HWS
HWS system Community level. The basic training (Text Document, Author:
involved standard operational Gordon Gow)
protocols, ICT maintenance/up-keep,
testing, and operational techniques.
Developing 03-Apr-2006 CAP Profile for Sri Lanka

protocols for (Text Document: Author:
warning & 05- Gordon Gow)
evacuation May-2006
Guidelines for HIH
(100%) (Text Document, Author:
Gordon Gow)
32

Education & TVEAP, the training partner, trained 03-Apr-2006 Awareness Training
awareness and certified 24 HazInfo Trainers. Modules 1 – 9
raising The HazInfo Trainers conducted 18-July-2006 (Text Documents:
awareness programs in the Authors: Buddhi
(100%) Community and at the District level. Weerasinghe, Nalaka
The HazInfo Trainers advocated the Gunawardene)
development of emergency response
plans (ERPs) for the Communities. Training List and
Assignments
(Spread sheet, Author:
Ravindra Kandage)
Plan and Program

Evaluation
Ravindra Kandage)
Assessment of Silent-Tests were conducted at the 01- Simulation Plans and
LM-HWS HIH, District, and Communities June-2006 Evaluation Report (Text
technologies between May-2006 and Document, Author: Nuwan
through drills October-2007 09- Waidyanatha, Peter
May-2007 Anderson)
(100%) Live-Exercises were conducted in the
Communities between Nov-2006 and
May-2007
Dissemination of Please see section 5.0 Project Output 12-Apr-2006 Last-Mile Hazard Warning
lessons learnt and Dissemination System in Sri Lanka:
throughout the 31- Performance of an Alerting
country Sarvodaya Media Department May-2007 and Notification Message
documented the activities at the HIH Relay
(95%) and the Communities during the (Text Document and
assessment of LM-HWS technologies Power Point Presentations,
during the simulated drills Authors: Nuwan
Waidyanatha, Gordon
Gow, Peter Anderson
HIH and Community

Simulation Video
Documentary (DVD,
Producer: Nishantha
Preethiraj)
33

Project LIRNEasia provided overall 01-Dec-2006 Project Interim Report 31-
management coordination and management May-2006
through the entire project 30-Nov-2007 (Text Document, Author:
(100%) Nuwan Waidyanatha)
Project Interim Report 30-

Nov-2006
Nuwan Waidyanatha)
Project Interim Report 31-

Mar-2007
Nuwan Waidyanatha)
Writing of final This document is the final report of 25- HazInfo Project Final
report with the project. May-2007 Technical Report
recommendation (Text Document, Author:
s 10-Jun-2007 Nuwan Waidyanatha)
(100%)
5. PROJECT DESIGN AND IMPLEMENTATION
5.1. Concept of Operations
The LM-HWS project consists of two information reception and dissemination stages:
• Reception and authentication of external hazard event and warning information by monitors lo-
cated at the Sarvodaya Community Disaster Management Centre (SCDMC) Hazard Informa-
tion Hub (HIH) located in Moratuwa (near Colombo) and dissemination to 28 (4 are controlled
villages) targeted Sarvodaya communities specially equipped with LM-HWS ICTs, and
• Reception and authentication of HIH generated messages by ICT Guardians in LM-HWS
equipped communities and dissemination of the messages to affected local populations.
When fully implemented, trained and certified monitors at the HIH will monitor, on an around-the-
clock basis, a variety of hazard and warning networks and disseminate alerts and other messages of
interest to LM-HWS equipped communities. There are 5 steps to initiating an alert message:
recording an event of interest, consulting with a Sarvodaya Executive, deciding to send the message,
composing the Common Alerting Protocol (CAP) message, and issuing the message. Alerts issued by
the LM-HWS are not “public” alerts and are intended only to be distributed to “first responders” who
34
have been trained and certified by the SCDMC as ICT Guardians within the LM-HWS Project.
The ICT Guardians are members of the local community and it is they, or their authorized designates,
who are responsible for determining if a local, community-wide (village) warning is to be issued. Upon
receiving an alert from the HIH, the ICT Guardian must first acknowledge receipt of the message to
the HIH and then proceed with notifying local community officials responsible for activating the
community emergency plan.
5.2. LM-HWS five wireless ICTs
The LM-HWS project is testing five information and communications technologies:
Dialog Remote Alarm Devices (RADs)
RADs are stand-alone units that incorporate remotely

activated alarms, flashing lights, a broadcast radio
receiver to be turned off or on and SMS messages to
be displayed, as well as self-test, message
acknowledgement and hotline GSM call-back
features.
Figure 9 Remote Alarm Device
GSM Java enabled SMS mobile phones (Java

phones)
Java phones receive text alerts in Sinhala, Tamil and

English (Java Phone), activate a Java applet to sound
an alarm, and hotline GSM call-back features.
Figure 10 Nokia6600 Mobile Phone
35
Disaster Warning Recovery and Response

Addressable Satellite Radio for Emergency Alerts
(AREA)
AREA is a WorldSpace satellite radio system that can

issue address hazard information directly to those
communities at risk. Global Positioning System (GPS)
technology incorporated into the radio receiver set,
along with the unique code assigned to every receiver,
Figure 11 AREA Receivers allows for hazard warnings to be issued to sets that are
within a vulnerable area or just to radio sets with
specific assigned codes.
Very Small Aperture C-Band Satellite Terminals

(VSAT).
VSAT terminals have been installed in two communities and

at the HIH. These facilities provide up to a 512 kbps Internet
connection and enable testing of the Internet Public Alerting
System (IPAS). An IPAS client application installed on a
computer enables pop-up messages to appear on a PC screen
and an audio alert tone to be played on the computer’s sound
system.
Figure 12 – VSAT System
CDMA Fixed Wireless Phones (FXPs)
FXP is a CDMA phones with built-in speakerphones to

provide voice communication via the public switched
telephone network.
Figure 13 - CDMA Fixed Phone
As illustrated in Table 2, each ICT relies upon a particular type of network access and message
management scheme.
36
Table 2 LM-HWS Access and Message Management Attributes

Message
Creation/disseminat
Device Network Access ion Interface Enabling Partner
RAD GSM fixed wireless Password protected Dialog Telekom Mobile
Internet website Communication
Laboratory at the
University
of Moratuwa, MicroImage
Java phones GSM mobile Password protected Dialog Telekom,
wireless Internet website MicroImage
Telephone CDMA fixed PSTN Sri Lanka Telecom
wireless
VSAT, Internet Internet via C-Band Password protected Innovative Technologies,
Public Alerting fixed satellite internet website Solana
System wireless Networks
Addressable L-Band portable Password protected WorldSpace Global Data
Satellite Radio satellite wireless Internet website Solutions
5.3. ICT Partners and their Contributions
Dialog Telekom: the largest mobile Telco and University of Moratuwa designed, built and supplied 7
Remote Alarm Devices (RAD), supplied 12 mobile phones, developed the Alerting software/applets
and provided a 128kbps microwave leased line for the HazInfo project. The “MIDewn” Java Applet
for mobile phones, was developed by Microimage (Private) Limited for Dialog Telekom.
Innovative Technologies installed VSAT satellite terminals that have proven to be a strategic tool that
can be used for high bandwidth TCP/IP communication for District Centers for VoIP, email, and
internet access during office hours and be available for Coordination of Recovery efforts from a
District Level. The same link can be the backbone for a Telecenter in the same location. The present
Sarvodaya organization communication depends on physical meetings and has proven to be a very
expensive process especially for Sarvodaya Staff in North-East parts of the country to travel to the
Head Office in Moratuwa (west-coast). A strategically placed set of C-Band VSATs can minimize the
travel time and accommodate Video Conferencing, technologies that are far more effective tool than
driving on A4 from Kalmunai to Colombo 362km one way.
Based on the post-Tsunami feedback from the affected countries, WorldSpace has designed emergency
alert system that incorporates CAP 1.1 formatted message generation, provision for auxiliary audio
delivery, automatic switchover at the remote sites, selective addressability, geo-referencing and a host
of other features that derive from a digital satellite radio broadcast system. The scheme has been
37
successfully field-tested and ready to be scaled up quickly. In addition to Sri Lanka, WorldSpace has
been testing their Addressable Radios for Emergency Alerts (AREA) products in India, Thailand, and
Indonesia. The effectiveness of the AREA system is the Audio and the Data Casting capabilities.
Major technical contributions came to the project via Canadian Experts such as Gordon Gow from
University of Alberta and Peter Anderson from Simon Fraser University who have been instrumental
in delivering the “Guidelines for HIH” that lays out the CAP Profile for Sri Lanka and the “ICT
Assessment and Community Observation Report” that identifies the shortcomings of the system,
respectively. Solana Networks, a Canadian Company, pro bono provided and managed IPAS services
for the field trials. This includes: full access to the Solana Networks IPAS server, support for creation
of public official, ensuring that alerts sent by public officials in Colombo are seamlessly and
successfully, upgrading and providing LIRNEasia with new IPAS client software technical support for
IPAS as required. Solana Networks intends to upgrade their IPAS system with improving direct access
to global sensor networks and adopting CAP for interoperability.
5.4. CAP GUIs and CAP Broker
Common Alerting Protocol (CAP) was integrated into the project because it is an open standard and
because of the perceived benefits and advantages of being an open source, XML-based protocol with
clearly defined elements and capable of supporting data interchange across multiple dissemination
channels. Ideally with CAP, one input at the central information hub can be translated into multiple
outputs for downstream alerting. CAP provides a standardized template for submitting observations to
the central hub (upstream) and thereby supports situational awareness to improve overall management
of a critical incident. Further, a CAP-enabled system will more easily integrate with other national and
international information systems (such as regional tsunami and weather alerting services).
An early challenge faced by the project was whether or not the project would need to build a CAP user
interface (GUI) from scratch, or whether there was an application available that could be borrowed.
The CAP GUI provides a method (a template) for authorized users to enter data into a computer at the
HIH and for the transformation of that data into the appropriate CAP XML elements to enable
standardized content creation and passage to the various ICTs. The CAP broker is a server application
that provides an intermediary point of interconnection between the Information Hub and the relay
network to facilitate interconnection of all ICTs and passage of CAP-compliant messages through a
single software application.
At this stage of the project, both WorldSpace and MicroImage have developed web browser based
CAP GUIs for the LM-HWS project. The WorldSpace CAP GUI (Anny Network Early Warning
System or Anny) interfaces with the AREA component and the MicroImage (Disaster Early Warning
Network or DEWNS) with the Dialog RAD/Java Phone components. Both the WorldSpace satellite
uplink and Dialog Telekom SMS servers are registered as teleports in their respective CAP alerting
software. Presently, the AREA and RAD/SMS CAP software operate independently of each other, but
in the future, could be integrated and controlled by a single software application (CAP broker).
Similarly, the VSAT IP gateway could be added as a teleport and integrated into a common CAP
solution along with other ICTs.
38
The VSAT satellite alerting message software also uses web browser based GUI called Internet Public
Alerting System (IPAS). IPAS in its current form is not CAP-compliant but provides a simple and
effective means to test sending IP based alerts to PC screens and audio systems.
Figure 14 DEWNS Browser SW Figure 15 ANNY Browser SW Figure 16 IPAS Browser SW
6. PROJECT OUTPUT AND DISSEMINATION
Project acquired a supplementary budget to disseminate the results regionally; the outcomes are
discussed in a separate report, which was submitted to IDRC along with this report. This section
documents events prior to the HazInfo dissemination work began.
6.1. Information Sharing and Disseminations
Event Description
Canadian Risk and Peter Anderson and Gordon Gow submitted a joint abstract presented
Hazards Network the HazInfo research results at the symposium. Nuwan Waidyanatha
Annual Symposium, was also invited to the event but was unable to attend the symposium
Vancouver 5-8
November 2007.
Pacific Northwest Anderson was invited to give a special presentation on the HazInfo
Preparedness Society project at the conference
Emergency
Preparedness
Conference,
Vancouver 20-22
November 2007
39
Event Description
Canadian SSHRC, HazInfo results will also be presented by Gordon Gow as part of the
Saskatoon, May 2007 Canadian Communication Association Conference at SSHRC Congress
in May 2007 in Saskatoon.
ISCRAM-2007, Delft, The paper co-authored by Nuwan Waidyanatha, Gordon Gow and Peter
Netherlands, May Anderson entitled, “Hazard Warnings in Sri Lanka: Challenges of
2007 Internetworking with Common Alerting Protocol” was peer reviewed
and accepted for Intelligent Human Computer Systems for Crisis
Response and Management, Delft, the Netherlands, May 13-16 2007
and to presented in the session titled – “Flexible Data Structures and
Services”. Nuwan Waidyanatha was invited to present the paper and
was granted a Euro 1000 Travel Grant to attend the conference.
Unfortunately was unable to attend as a result of not receiving a visa to
travel to Netherlands on time.
ITU-Arabia, The International Telecommunication Union (ITU), the League of Arab

Alexandria, Egypt, States (LAS) and a number of other United Nations Agencies in
April 2007 cooperation with the Arab Academy for Science & Technology &
Maritime Transport (ITU-D sector member) are organizing a Joint
Regional Conference. The event is on “Disaster: Relief and
Management: International Cooperation & Role of ICT”, and will take
place in Alexandria, Egypt, from 14 to 17 April 2007.
Journal of Emergency A paper authored by Gordon Gow based on use of CAP for the HazInfo
Management, March/ project titled – “Implementing Common Alerting Protocol in Sri
April 2007 Lanka” was accepted and published in the Journal of Emergency
Management, Volume 5, Number 1, pages 50-56.
Institute for The Media Conference is following an open public forum, and is being
Construction Training held to enlighten the media and the General Public of the progress
and Development made on disaster warning with a focus on community based last-mile
(ICTAD) warning system. All key stakeholders of the project will attend and will
“Savsiripaya”, engage with the press in educating them with the lessons learned.
Colombo, Friday 30th
March, 2007
Sarvodaya Head Forum: “Sharing Knowledge on Disaster Warning, with a focus on

Quarters, Moratuwa, Community-based Last-Mile Warning Systems” brought together
Wednesday 28 & knowledge from the South Asia Region to Sarvodaya for a forum on
Thursday 29 March, practical solutions for communicating risk information to rural
2007 communities; learn about end-to-end hazard detection and alerting
systems that serve grassroots communities in the South Asia Region;
obtain feedback on the outputs of “Evaluating Last-Mile Hazard
Warning Dissemination: A Research Project” and the Sarvodaya
Community Disaster Management Center “Public/Private Partnership
40
Event Description
Information Communication Framework”.
Improving British Experience in applying Sri Lankan ‘last mile’ warning techniques is
Columbia Tsunami being shared with coastal villages along the Pacific coast of B.C,
Warning, Haida Gwaii Canada, and within the B.C. Tsunami Integrated Preparedness Project
Communities, Queen (TIP), the main multi-stakeholder tsunami planning group. Prof.
Charlotte Island in Anderson provides regular updates to TIP on progress of the HazInfo
North Pacific, January project. In January 2007, just before returning to Sri Lanka, he gave
2007 presentations and conducted consultations in seven Haida Gwaii
communities (Queen Charlotte Islands in northern Pacific) on
emergency communications and tsunami warning drawing upon on
lessons learned from the HazInfo project.
ITU-ESCAP, Bangkok, The conference was aimed at addressing interoperability between

Thailand, December communication systems, public network utilization for emergencies,
2006 required enhancements for public infrastructure, and recent technology
changes. Participants also defined the requirements for public warning
as well as identifying possible solutions/features of technologies to be
used for emergency telecommunications. Prof. Peter Anderson (Simon
Fraser University) and Dr. S. Rangarajan (WorldSpace) attended the
sessions and presented the HazInfo project and the AREA solutions at
the conference.
CBC Radio One, Professor Rohan Samarajiva and Dr. Gordon Gow were interviewed on
Edmonton, Canada, CBC’s “As it Happens” program.
December 26, 2006
Bridging the Long Media Development journal has published Mr. Nalaka
‘Lat-Mile’ in Sri Gunawarndene’s essay titled Bridging the Long 'Last Mile' in Sri Lanka
Lanka3: Media in its latest issue (2006/4) which is thematically focused on disasters
Development Journal, and communication. This is a non-technical description of what we are
UK, 17 November, trying to do in the HazInfo Project. Nalaka Gunawardene is the Chief
2006 Executive Officer of TVE Asia Pacific (Private) limited a lead partner
in this project. TVE Asia Pacific conducted the Training-of-Trainers
program for the HazInfo project.
ITU/OASIS Workshop Dr. Gordon Gow, Communications Consultant for the HazInfo project
and Demonstration on presented the CAP initiative in Sri Lanka to the International
Advances in ICT Telecommunications Union in Geneva. This paper presents the
Standards for Public preliminary findings that involve the implementation of the CAP
Warning: standard to support a local all-hazards warning system in Sri Lanka. In
particular, it describes the challenges of implementing a CAP-based
3
The paper can be found here --
http://www.wacc.org.uk/wacc/publications/media_development/2006_4/bridging_the_long_last_mile_in_sri_lanka
41
Event Description
information system for managing multi-lingual warnings across a set of
five technologies in 32 tsunami-affected villages along the southern and
eastern coast of Sri Lanka.
Coastal Community Nuwan Waidyanatha presented the HazInfo Project initiative at the
Resilience Initiative: USAID sponsored forum. Under the USAID-funded US Indian Ocean
USAID, TAJ Hotel, Tsunami Warning System (IOTWS) Program, Coastal Community
Colombo, 20 Resilience (CCR) is a focused initiative that promotes tsunami and
September, 2006 other hazard readiness through active collaboration of national and
local emergency management agencies, coastal managers, training
institutions, and local communities. Building on local knowledge and
needs, this effort supports integrated and more standardized hazard
awareness and mitigation efforts that improve public safety during
emergencies and builds preparedness against recurring disasters.
Institute of Policy This workshop was conducted to bring to together stakeholders and
Studies Sri Lanka and share the findings of the review of “Disaster Management: Policy and
Oxfam America Practice” undertaken by the Institute of Policy Studies (IPS) of Sri
Disaster Management Lanka with financial and logistical support from Oxfam America. The
Forum: TAJ Hotel, IPS report is intended to assist the government, NGOs and other
Colombo, 17 stakeholders to improve disaster management in Sri Lanka. It is just
September, 2006 over a year since the Parliamentary Select Committee on Natural
Disasters presented their recommendations and it seems timely to
review whether these have been implemented in practice.
Using ICTs for The Commonwealth Telecommunications Organization of UK held a 3

Disaster Management: day conference in Ahungalle, Sri Lanka, which brought together
Heritance Hotel, predominantly Asia Pacific Region and a few International delegates in
Ahungalla, 17 June, the field of Disaster Management to present the ongoing projects and
2006 initiatives for risk reduction. Rohan Samarajiva was the Keynote
speaker and Chair for day one. Nuwan Waidyanatha presented the Last-
Mile Hazard Warning Dissemination: A Research Project.
Protocol (CAP) The purpose of this discussion was to provide a common point of
Teleconference for a reference for those parties that will be meeting to discuss the
Last Mile-Hazard implementation of Common Alerting Protocol within the
Warning System in Sri Sarvodaya/LIRNEasia “Last Mile Hazard Warning System” (LM-
Lanka, 22 March 2006 HWS) project. The scope confined to the implementation of Common
Alerting Protocol (CAP) for the project.
Planning workshop on LIRNEasia issued invitations and ensured participation by all partners
Last-mile Hazard other than Sarvodaya. Sarvodaya, being the provider of the location,
Information Project: took charge of the organizational responsibilities; arrange the actual
Sarvodaya HQ, meeting location, interpretation, etc. LIRNEasia ensured background
42
Event Description
Moratuwa, 22 material was distributed beforehand. The program ensured that all
January, 2006 partners have common understanding of the project; achieve common
understanding of the ethical issues and our approach; obtain fullest
participation by and input from partners on the actual work plan of the
project. A task matrix with deadlines and responsibilities were prepared
and clearly agreed upon by all partners engaging in the LM-HWS
project. Please see Annex 1 – Planning Workshop Program.
6.2. Knowledge Creation
Event Description
Public Lecture: Sri The lecture was free and was open to the public. The lecture with title:
Lanka Foundation Responsive Innovation for Disaster Mitigation addressed all-hazards
Institute, Colombo, 20 warning and the use of the Common Alerting Protocol in disaster
July, 2006 mitigation. Gordon Gow is an Assistant Professor in the Faculty of
Extensions at the University of Alberta, Canada. Co-author of the book:
“Mobile and Wireless Communication: An Introduction” and most
current book: “Policymaking for Critical Infrastructure”. Moreover, he
is the communication systems consultant for “Evaluating a last-mile
Hazard Dissemination: A Research Project” in Sri Lanka.
University of Wash- A special international certificate program in tsunami warning is being

ington Extension De- designed for and offered to emergency managers and policy makers
partment – from the Indian Ocean Region by a consortium of international experts
USAID/NOAA Tsuna- and key warning agencies under the coordination of NOAA. Prof. An-
mi Certificate Pro- derson is a member of the Certificate Program Advisory Board and an
gram, July 2007 invited instructor for the Program’s first course offering in July 2007.
He will be leading a special session on ‘last mile’ warning techniques
and challenges and will base his presentation largely on the Sri Lankan
experience.
International Prof. Gordon is currently working with ITU-D (Development Sector)
Telecommunications on a set of international guidelines for public alerting that will include
Union (ITU) an entire section devoted to results from the project in Sri Lanka.
6.3. Training
Event Description
Sarvodaya HQ, All ICT Guardians, HazInfo Trainers, District Coordinators, and HIH
Moratuwa, 17 staff participated in a workshop to discuss the procedures for the Live
November, 2006 Simulations. The Workshop was chaired by Prof. Peter Anderson
43
Event Description
(School of Communication, Simon and Frasier University). Prof.
Anderson is the consultant monitoring the simulations and evaluating
the HIH. HazInfo Trainers were educated on how to lead a team in
observing the community simulations, collecting data, and preparing
the simulation evaluation reports for each of the communities. District
Coordinators are to oversee the simulation exercises of the
communities belonging to their districts. The Divisional Coordinators
will assist the District Coordinator and take the responsibility of
formally informing the local Government, Police, Medical Officer
Health, and Gov Disaster Management Officials.
HazInfo Trainer Shanthi Sena of Sarvodaya organized for the HazInfo Trainers to gather
Refresher Course: at the Head Quarters to get a refresher course on Training Villages. At
Sarvodaya HQ, the same time the HazInfo Trainers were motivated to by providing an
Moratuwa, February incentive scheme to complete the Simulations. Dr. Buddhi
2007 Weerasinghe, Ms. Natasha Udu-gama, and Nandana Jayasinghe made
presentations and conducted discussions to enhance awareness.
Training-of-Trainers This was a residential workshop where accommodation was arranged

Workshop at for all participants, and some resource persons, at the venue itself. All
Sarvodaya in meals were provided. The entire training event was conducted in an
Bandaragama, Sri informal, yet structured manner. Participation in all sessions was
Lanka, 03 April 2006 mandatory for all participants. “Lecture Presentation”: This involved a
PowerPoint presentation by resource person/s, sometimes supported by
video clips. There was time and opportunity for discussions in all such
presentations. “Group work”: Most sessions had a specific time for
participants to divide themselves into 3 – 4 groups to engage in parallel
activities as briefed by the resource team. “Group presentations”: This
harnessed a short, summarized presentation by the groups through a
member of each group selected as rapporteur. “Video presentations”:
These carefully selected video films integral to the workshop content
were shown; preceded and followed by targeted discussions by the
resource team. Synopses of videos given towards the end of this note.
Common Alerting Awareness to village Information and Communication Technology

Protocol Workshop guardians and District Coordinators of each Information and
Part II: Communication Technologies are evaluating in Last Mile Hazard
HQ, Moratuwa, 13 Warning System (Research Project).
September, 2006
Hazard Information This is a course for the training and certification of the hazard
Hub Monitor Training Information Hub Monitors. Basis of the course is the Common Alerting
and Certification Protocol4 (CAP) for Risk Information Communication. The alert
Course: Sarvodaya messages are issued by an “authorized user” through the Sarvodaya
44
Event Description
Disaster Management Disaster Risk Management Center Hazard Information Hub (HIH). By
Center, Moratuwa, 10 the end of the course the Monitors were able to -- 1) Identify an Event
August - 20 of Interest; 2) Confirm EOIs with an Executive; 3) Construct a CAP
September, 2006 Message 4) Relay the CAP message. Content emphasized, in the event
the government of Sri Lanka issues a public warning, how the Monitor
will relay this message directly through its network. If the government
does not issue CAP-compliant messages, the Monitor will need to
develop a system for translating or reformatting these messages quickly
and accurately.
Common Alerting Ensure that all participants, namely Sarvodaya First Responders (ICT-
Protocol Workshop I: Guardians), have a good understanding of the Common Alerting
Sarvodaya HQ, Protocol. Provide ICT Administration and User training to all Village
Moratuwa, 21 July, and District First-Responders. Introduce the Hazard Information Hub
2006 Help Desk functions to the participants. Give the technology partners
who designed and developed the ICTs an opportunity to demonstrate
their solutions. Handover the ICTs to the Village First-responders.
Update the Sarvodaya District Centers of the Last-Mile HazInfo Project
current status.
Addressable Satellite Dr. S Rangarajan (Vice President Engineering) and Dr. Wilson Baker
Radio Training by (Vice President International Projects) conducted one-day training on
WorldSpace: standard operational procedures (2 person authentication rule),
Sarvodaya Disaster configuring DAMB-R2 Receivers, and using ANNY Network software
Management Center, to send messages.
Moratuwa, 07 June,
2006
4
http://www.incident.com/cookbook/index.php/Welcome_to_the_CAP_Cookbook
45
7. CAPACITY BUILDING
7.1. Local Research and Development
On the ICT development side, the project significantly helped to strengthen and mobilize local
research capacity by working with local laboratories such as Lanka Software Foundation (LSF),
Dialog-University-of-Moratuwa Communication Research Laboratory and Microimage.
LSF is the advocate of “Sahana”, meaning Relief. Sahana is a Free and Open Source Disaster
Management Software developed Disaster Management Practitioners in the World and a core Software
Engineering Team of the Lanka Software Foundation based at the University of Colombo. The Sahana
Team developed the Messaging Module for Sahana and included the CAP component in there. A
Demo5 of the current version is deployed in the demo release of Sahana.
7.2. Hazard Information Hub Monitor Course
In August 2006, HIH staff were given training for certification as Hazard Information Hub Monitors
and “authorized users” (HIH Monitors). The core element of the course used the Common Alerting
Protocol for supporting message creation and communication. By the end of the course, the HIH
Monitors were able to: 1) identify an Event of Interest (EOI), 2) confirm the EOIs with a member of the
Sarvodaya Executive, 3) construct a CAP Message, and 4) relay the CAP message to designated
communities using a particular combination of LM-HWS ICT devices.
The training also emphasized, in the event the government of Sri Lanka issues a public warning, how
the HIH Monitor should relay this message directly through the LM-HWS network. If the government
does not issue CAP-compliant messages, the Monitor will translate or reformat these messages quickly
and accurately into a LM-HWS CAP-compliant message.
Technical Annex 15 contains the training material and the procedural guidelines for the HIH.
7.3. Village Training of Community First-Responders
Training at the village level consists of two training components: 1) general disaster preparedness
training for village trainers and 2) specific training for ICT Guardians.
Training of Trainers
The Training of Trainers (TOT) component forms the education component envisaged in LM-HWS as
the underpinning training to instill an effective warning response. Those trained (known as HIH
Trainers), in turn, are to train ICT Guardians and others at the community level.
5
To access Sahana Demo -- http://demo.sahana.lk/
46
The newly trained trainers are expected to be able to:

 Instill awareness of hazards, vulnerability, and risk.
 Conduct community based hazard mapping, resource mapping, risk assessment and identification
of vulnerable households and groups of people in the community.
 Create awareness of the early warning mechanism in place and the response activities that need to
be undertaken.
 Enable the identification of appropriate dissemination mechanisms to vulnerable households.
 Enable formulation of a response plan, which includes an evacuation plan with clear actions, roles
and responsibilities and identify resources needed.
 Instill the competency to conduct an evacuation drill which is comprehensive.
 Provide the ability to measure the effectiveness of response plans carried out as simulations.
TOT took place in Bandaragama in April 2006 and February 2007 and was conducted by the training
partner – TVE Asia Pacific. Currently, 16 of the 32 participating communities received training as part
of a controlled group in the pilot study. Technical Annex 16 – 23 contains the course material used in
the 8 training sessions.
ICT Guardian Training
In July, 2006, a workshop was held at Sarvodaya in Moratuwa to train the Sarvodaya Village and
District first responders with the objectives of:
 Ensuring that all participants, namely Sarvodaya first responders, have a good understanding of the
Common Alerting Protocol.
 Providing ICT administration and user training to all village and district first responders.
 Introducing the Hazard Information Hub Help Desk functions to the participants.
 Giving the technology partners who designed and developed the ICTs an opportunity to demon-
strate their solutions.
 Handing over the ICTs to the Village first responders. Those responsible for their operation were
designated as “ICT Guardians”.
This training was supplemented through district level workshops conducted prior to community
simulation exercises from November 2006 through March 2007.
Another change in the project was the decision to Train the Untrained Communities. This would
happen shortly after the completion of simulations. This way it wouldn’t affect the research results and
all participating villages will be trained to the same level and all villages will have improved their
alerting and response capacity.
47
8. PROJECT MANAGEMENT
The Last Mile Hazard Information Dissemination Project was a multi-partner, civil society initiative to
complement other action being taken at national and regional levels. However, both private sector and
public sector participation was under-represented. Only WorldSpace and TVEAP took a large interest
in the project; all others gave a lower priority to the project. Hence, there were delays in activities,
additional expenditures of time and resources to achieve goals, and overall a lower level of outcome
relative to expected results. The media partners (except TVEAP) had failed to deliver. The VSAT
partner was slow in delivering; the project had no choice but to use the exclusive vendor available in
the market, and to bow to the vendors’ time line. Although, this turnkey component was slower in
being inaugurated, the quality of service of the VSAT provider has been excellent.
The Sri Lanka “North-East” Conflict between the Liberation Tamil Tigers for Elam and the Sri Lanka
Government has forced the project from running Live-Exercises in Jaffna District. Similarly, the
Hambantota District was affected by floods and damn breaks, giving people enough to worry about
than concentrate of Live-Exercises. The drills were completed later in May 2007. At least 1 Silent-Test
with the ICT-Guardian was conducted by the HIH-Monitor initiating the Silent-Test. Table 6 below
defines the allocation of ICTs and the participating communities with the specific ICT.
Project Management in Sri Lanka requires plenty of buffer time because of anticipated delays. The Sri
Lankan society is not used to working according to plans, nor do they adhere to concept of deadlines.
The reason being Sri Lankan society is used to giving excuses for not being able achieve the targets
opposed to concentrating on the tasks and striving to provide solutions. As a result the 2-year project
was re-planned with multiple parallel tasks over a 1-year period. The outcome of this tight-plan was
that the project completed all activities in 18 months.
The Project Plan was constructed anticipating all tasks would be delivered on time. Therefore, the
Training began immediately following the planning meeting anticipating that the ICTs would be
delivered and commissioned on time. However, this assumption backfired causing a huge gap between
the training period and the delivery of equipment. Therefore, the communities had to be given a
refresher cause after all the ICTs were delivered. The delay in ICT delivery caused the simulations to
be delayed as well. The communities had been trained at the beginning of the project and they had
developed their ERPs at that time. Because the simulations happened only 8 months after, most
communities had forgotten their emergency response plans. In the future, it is important to wait for the
delivery of equipment and services before providing the training at the last moment.
Originally the Sarvodaya District Coordinators and Division Coordinators were trained at workshops
to take the training in to the communities to ready them for the simulation activities. However, this
strategy failed and multiple field visits had to be made to motivate and train the communities to engage
in the simulations. Also the Sarvodaya District Coordinators, Divisional Coordinators, and Volunteers
had to be given monitory incentives in order to motivate them to conduct project activities. The
incentives were based on the delivery of community simulation progress report, which consisted of
assessments provided to them by the project.
48
The original budget of the project had to be restructured to fit the actual requirements. Otherwise, it
would have been impossible to take the project forward. The major components were reallocating
funds for much needed Workshops for capacity building and reducing the number of VSAT stations.
Dialog Telekom, GSM technology Partner, donated time and equipment. Therefore, the project was
able to reallocate money set aside for this equipment towards Workshops. The number of VSATs had
to be reduced because the money allocated in the original budget was for Ku-band VSATs, which have
proven to be very unreliable in the tropical climate of Sri Lanka. Therefore, slightly more expensive C-
band VSATs had to be used in the project. Also Infrastructure, Office-equipment and IT Hardware that
were allocated for all 5 VSAT Stations were reallocated towards Workshops because the project was
able to use IT Hardware and Infrastructure that was already available for Sarvodaya through other
projects.
Simulations were first scheduled to be conducted over a 6 week period starting in November and
Finishing in December. This strategy failed because the Sarvodaya District Coordinators could note
agree to the dates due to commitments to other project activities. North-East (NE) District
Coordinators were tied up with the Internally Displace People (IDP) as a result of the NE conflict. A
couple of the Communities and District Centers had Organizational dysfunctional issues. The HIH did
not have adequate capacity to assist in deficiencies in scheduling and executing simulations in the
communities.
9. IMPACT
Last-Mile Communities realize the potential of the LM-HWS to help them sleep easy at night.
Revoking it without enhancing it based on the research findings and also expanding into the remaining
15000+ Communities in Sri Lanka would jeopardize the trust they have in Sarvodaya. The
communities strongly requested that the DM programs be continued and not let them subside as most
things do. All Sarvodaya trained villages had started an “Emergency Disaster Management
Committee”. In most cases the committee comprises members of the Sarvodaya Shanthi Sena
Volunteers. Table top exercises revealed that all communities needed guidance and assistance to
strengthen the resilience in their neighborhoods.
Sri Lankan Government is only now firming the mechanisms of early warning and dissemination and
the experiences of the Last Mile HazInfo Project may be invaluable in this process in the coming years.
The research aims of the project were not always apparent to those involved at the grassroots level
delivery of project services or to the beneficiaries themselves. This led to unrealistic expectations in
some cases, for example when certain communities wanted the VSAT rather than an ordinary fixed
phone or mobile phone. A common psychology among communities is to regard a more complex (and
expensive) technology tool as somehow more important than a ubiquitous one.
The disadvantages encounter in the project may convert itself into strengths in the future as lessons
learnt at community level would certainly enrich the next phase of the project that is being envisaged.
49
A critical drawback of the project was that most of these village-trainers and HIH Monitors could not
be retained with the project due to various personal factors. As such, some left when they gained paid
employment. The second batch of volunteers, although given some orientation and guidance, did not
receive the same intensive training that was directed to the first batch. This caused a loss of some mo-
mentum and motivation, but it was unavoidable. Such a level of turn over in personnel may cause an
imbalance in the integrity of the system. As a result the LM-HWS may not be able to provide the ex-
pected level of services the communities require.
10. OVERALL ASSESSMENT

In general, the project revealed that with the recommended enhancements described in this report, the
five tested ICTs can be incorporated into the communities and form a critical infrastructure. However,
most importantly, from a technical perspective, all ICTs used in the HazInfo Pilot must be upgraded to
receive Complete Full-CAP Messages before they can be used in the Last-Mile Communities of Sri
Lanka. Further, before the CAP Profile can be implemented in such a last mile system, the system must
first develop the Human Capacity: HIH-Monitors, ICT Guardians, and ERP Coordinators, in order to
supplement the deficiencies of an end-to-end fully-automatic early warning system. Thereafter,
simulated drills must be conducted regularly to develop the Cognitive Framework to ensure all ERPs
can be smoothly carried out without confusion.
From the simulations is apparent that given proper training of HIH staff, timely access to external
hazard event information and the appropriate ICT, the time taken to process and disseminate an alert
from the Hazard Information Hub (HIH) can be absolutely minimal. This can give the communities,
the majority of the time (75% or more between the time of receiving the event information at the HIH
and the estimated time of hazard impact) to execute their ERPs. However, the project sees that this
timing advantages can only be effectively achieved by a “CAP Broker” that will integrate and improve
interoperability among the ICT CAP systems and provide the single input - multiple output facility that
HIH Monitors need to speed up the capturing of event information, get authorization and disseminate
CAP messages efficiently and accurately. The CAP Broker can be developed by redesigning the
existing tested WorldSpace ANNY, Dialog-University-of-Moratuwa-Microimage DEWNS, and
Solana Network’s IPAS software systems. In addition the CAP Broker would require a Geological
Information Systems (GIS) based Graphic User Interface (GUI).
A disadvantage was that the absence of a culture of last mile dissemination of early warnings in Sri
Lanka – in this sense, HazInfo project was breaking new ground. The practice until now has been for
information to be ‘broadcast’ to the whole country in a central manner. Introducing Addressability and
the community-based approach was a challenge, and gathering momentum in the field was at times
slow due to this reason.
50
11.RECOMMENDATIONS
11.1.Message formatting - the HIH and the Government
In the event the government of Sri Lanka issues a public warning, the HIH will relay this message di-
rectly through its network. Local first responders will act on the message as they see fit, based on local
response plans and any instructions provided by the government in the initial message or in subsequent
public communications (e.g., official messages broadcast by the media).
If the government does not issue CAP-compliant messages, the HIH will need to develop a system for
translating or reformatting these messages quickly and accurately. If government messages are issued
in another standard format it might be possible to automate this process. However, in no case should
personnel at the HIH modify or otherwise revise the contents government message, except to ensure
that it is capable of being relayed over the HazInfo network.
It is recommended that an element in the CAP data definition document (the Profile Document) be as-
signed to indicate that a warning message has been issued directly from the government of Sri Lanka.
In fact, it is advisable that the selected CAP data element be able to contain the entire government mes-
sage as a single payload that can be copied and pasted into a CAP message, rather than asking HIH
personnel to interpret or to transcribe the message and risk introducing errors in the transcription pro-
cess.
11.2. Further Test ICT Assessment Tools and Methodologies
In addition to the core activities, the Partnership for Public Warning (PPW) has identified a list of rec-
ommended best practices (RBPs) to consider when devising a warning strategy. These RBPs can be
applied to each of the three core activities to produce a matrix with specific recommendations for the
HazInfo initiative. The first column on the left lists 39 RBPs based on those identified by the PPW.
The second column lists operational issues related to the RBPs. The third column lists approximately
120 specific recommendations for the design and implementation of a HazInfo system. These are cate-
gorized into each of the three core activities: planning, education and training, testing and evaluation.
11.3. Table Top Exercise for Selected Members in Community
It is recommended that all villages participating in the project benefit equally from training
opportunities offered by the HazInfo Project. After consultation with HazInfo Project Districts and
Villages at the pre-simulation workshop, consensus was reached that it would not be appropriate to
fully exercise villages that are not trained, to avoid creating confusion among residents about the
intended purpose of the simulation. Further, there are public safety considerations that must be
considered in advance whenever public participation is sought in such exercises. There was a change
of strategy to engage untrained villages in the simulation exercise. Although the simulation exercise
51
proceeded as originally planned and untrained villages did not receive additional training up to the
time of receipt of the alert message from the HIH, at each untrained village HIH trainers and District
coordinators conducted a “table top exercise”; with relevant officials at the village location where the
Last Mile alerts would arrive. The table top exercise was in the form of an around-the-table discussion
about what steps the village would take in the event the alert message was real, followed by HIH
trainers taking participants through an introduction to disaster planning; to be followed by a full
training workshop at a future date. In this way, shortly after the completion of simulations all
participating villages would be trained to the same level and all villages have improved their alerting
and response capacity.
This procedure provides a model for future expansion of the project. A slightly modified version of
this table top process can also be applied in trained villages where circumstances prevent them from
conducting full community simulations. All villages, trained and untrained, will also be able to
complete the HIH-to-village Last Mile technical simulations.
11.4. Training and Certification of HIH Monitors
It is recommended that an authorization procedure be developed for HIH staff members to qualify
them as “Authorized Users” with the authority (and related responsibilities) to compose and issue
warning messages over the HazInfo system.
All new HazInfo personnel should be required to attend a training session provided by the
HIH/Sarvodaya. In addition all authorized users should be required to attend regular refresher training
to maintain their status. Regular training sessions could be scheduled prior to the cyclone season in
November. In addition, authorized users should be required to complete monthly practice sessions
composing and initiating messages using the system. To promote compliance, the HIH computers
could be designed to log these monthly sessions.
11.5. Training of ICT Guardians
It is recommended that HIH develop a series of basic HazInfo set up, operational and maintenance
checklists and guidelines, and that they be distributed in all appropriate languages. It is further
recommended that HIH examine how HazInfo components could be used to deliver regular updates to
ICT Guardians – especially the AREA audio service and that refresher and advanced user courses be
considered as well as Training Of Trainer program for ICT Guardians be offered to encourage inter-
community self-help.
11.6.Community-based Emergency Communications Planning and Training
It is recommended that a new module be added in the next phase of the project that would include
community-based emergency communications planning incorporating communications needs
assessments based upon community emergency response plan concepts of operation, an inventory of
52
communication means, communications capacity and vulnerability assessment and gap analysis, along
with operational training.
11.7.Contribution to CAP Research through Working Groups
It is recommended that a special R&D working group be established and retained to pursue further de-
velopment of CAP and its integration with the project’s and other emerging ICTs that can support haz-
ard warning and situation awareness.
A major component of the project is the use of the Common Alerting Protocol to enable data inter-
change between the HIH and a range of technologies. However, there are many issues remaining to be
resolved in order to fully operationalize CAP. These include interoperability, language and message
display capacity and are further discussed below. Because of the global characteristics of these imple-
mentation issues, the CAP component of the study could make an important contribution regionally
and perhaps worldwide for the deployment of this relatively new standard for emergency alerting. As
such, considerable attention should be paid to the CAP implementation methodology and to documen-
tation of the implementation process.
Some decisions are also concerned with localizing CAP for the specific Sri Lankan context (e.g., lan-
guage, geographical regions, hazards, etc.). Other decisions are concerned with information manage-
ment requirements that stem from the use of CAP and the long term path dependencies associated with
future expansion of CAP within Sri Lanka and in the Indian Ocean region.
The decision-making process is also an important component of the pilot study, as it will provide
lessons learned for implementing CAP in other organizations in Sri Lanka and in other countries. For
this reason, it is important to preserve a record of events during the implementation and testing stages
and to encourage broader participation through a working group.
11.8.Enhance and Test ICTs with Complete and Full CAP Features
It is recommended that all HazInfo ICTs be capable of full CAP featured and multilingual message
display. In this project, a Complete Full-CAP Message is defined as one that complies with the CAP
Profile for Sri Lanka, contains all three languages: Sinhala, Tamil, and English, and also is
disseminated in modes of Audio (i.e. Voice) and Text. The final rating is the multiplication of the
values obtained from Tables 2, 3, & 4.
Presently, the HazInfo project does not employ a multi-langauge device that is capable of displaying
the entire CAP message. The WorldSpace optional Alert Box is the only device that can display the
entire CAP message. However, it is currently configured for English language character only.
Therefore, ICT-Guardians have had to learn to interpret the English partial CAP Messages.
Considering the low level of English language usage especially in rural Sri Lanka, ideally, the devices
should enable the ICT Guardians to select the message language of choice (i.e. Sinahala, Tamil, or
53
Englsh).6
Of the five ICTs used in the project, only three make use of the CAP message format (AREA, RAD
and MOP). However, each is only capable of displaying a limited number of the CAP elements,
limiting the amount of alert message content. Consequently, during simulations ICT-Guardians
restricted their CAP message content creation to recording only a few elements such as the
<msgType> and <Event> and did not include <urgency>, <severity>, and <certainty> to enable ICT-
Guardians to gage the Priority. One reason particular reason for doing so was that only the WorldSpace
AREA units could display these qualifier elements. Although, the RAD and MOP devices use the
<Description> element that could also carry this information, they are restricted to an overall message
content limit of 130 characters.
11.9.Free and Open Source Software CAP Broker
It is recommended that a free and open source software program (CAP Broker) be developed to
integrate the project’s individual CAP message generators and processors to serve as a single non-
repetitive data input and import/export function.
An early challenge faced by the project was whether or not the project would need to build a CAP user
interface (GUI) from scratch, or whether there was an application available that could be borrowed.
The CAP GUI provides a method (a template) for authorized users to enter data into a computer at the
HIH and for the transformation of that data into the appropriate CAP XML elements to enable
standardized content creation and passage to the various ICTs. The CAP broker is a server application
that provides an intermediary point of interconnection between the Information Hub and the relay
network to facilitate interconnection of all ICTs and passage of CAP-compliant messages through a
single software application.
At this stage of the project, both WorldSpace and MicroImage have developed web browser based
CAP GUIs for the HazInfo project. The WorldSpace CAP GUI (Anny Network Early Warning System
or Anny) interfaces with the AREA component and the MicroImage (Disaster Early Warning Network
or DEWNS) with the Dialog RAD/Java Phone components. Both the WorldSpace satellite uplink and
Dialog Telekom SMS servers are registered as teleports in their respective CAP alerting software.
Presently, the AREA and RAD/SMS CAP software operate independently of each other, but in the
future, could be integrated and controlled by a single software application (CAP broker). Similarly,
the VSAT IP gateway could be added as a teleport and integrated into a common CAP solution along
with other ICTs.
The VSAT satellite alerting message software also uses web browser based GUI called Internet Public
Alerting System (IPAS). IPAS in its current form is not CAP-compliant but provides a simple and
effective means to test sending IP based alerts to PC screens and audio systems.
6
A report by Anderson, Peter [1] describes in detail the shortcomings of the ICTs and recommends enhancements to make the ICTs
usable in the last-mile of Sri Lanka. The report further proposes an integrated software system that can automate some of the HIH
processes to make it easy on the HIH Monitors to minimize the dissemination times.
54
Global hazard monitoring organizations such as United States Geological Survey (USGS) and Pacific
Tsunami Warning Center (PTWC) also issue CAP format messages. However, throughout the project,
it was discovered that not all organizations, including USGS and PTWC, use the same OASIS CAP
version making it difficult to automatically import their CAP Messages into the HazInfo CAP readers
and thus requiring manual transfer of data for relaying timely Events of Interest.
For the HazInfo CAP Broker, the current inputs will come in the form of text and voice. The message
editor should be capable of transforming the audio and/or text to fit the defined CAP format. In most
cases the message will be entered by the user. In a few cases the interface to external sources will be
direct without any user interference.
A natural language translator will take a CAP input in any language and translate the message to select
other languages. The user should be able to switch between languages as and when needed. Hence, the
GUI must also change to fit the localization requirement; i.e. switch to the selected language. The user
can then enter the CAP message in their language of choice.
Thereafter, the translator in the Message Relay component will convert the message to the specified
languages and file formats.
After the messages are generated using the editor and the CAP message is saved, then the Relay
component will generate the necessary and sufficient outputs in the specified file formats. Thereafter,
these files will be relayed to the registered gateways. In most cases it will be a TCP/IP file transfer that
will be picked up by the provider. The module will provide means to configure the message relay
component. This will entail mapping the CAP message file formats with the respective Gateways.
Hence, this component will also allow the user to register and configure the gateway (teleport). The
providers will deliver the messages to the last-mile.
11.10.Closed User Group Digital Audio Broadcasting
It is recommended that Sarvodaya explore the feasibility of establishing its own national radio network
as a value added programming service to the WorldSpace component of the HazInfo Project.
This project highly emphasizes “Early Warning” stage communications. However, the project also sees
importance in the HIH being able to function before and during response and recovery stages of
disasters. A number of the ICTs rely upon terrestrial infrastructure backbone networks that could be
impacted by hazards. Two ICTs provide alternate means to communicate between the HIH and
villages and are not dependent upon terrestrial networks. These are the expensive VSAT system and
the WorldSpace satellite systems. Of particular interest are the WorldSpace audio broadcast
capabilities. The project has tested a digital radio channel for normal programming including test
transmissions in Sinhala, Tamil, and English with instructions that can be easily recorded and
transmitted via the project’s own, “Sarvodaya Talk” 950 World Space audio channel.
Initial tests indicate that it’s feasible for Sarvodaya to operate the “Sarvodaya Talk” audio channel on a
24/7 basis; particularly for communicating to villages during response and recovery phases of a
55
disaster. The difficulty is absorbing the high cost of the WorldSpace channel. However, if Sarvodaya
can devise an appropriate business plan, it may be able to recover the cost from the various Sarvodaya
Units such as “Information Technology Unit”, “Economic Enterprise Development Services”, “Shanthi
Sena”, “Suwadana Centers” and other Sarvodaya programs by charging a nominal fee to broadcast
news and their educational and other programs directly to the communities based on fixed weekly-
program. For example, the SNeHA Project is an effort to strengthen health services in Sri Lanka by
establishing ICT infrastructure for eHealth and enhancing capacity of healthcare workers in over a
dozen of the districts. The project used a WorldSpace solution, namely the AREA-A with datacasting,
in its eLearning Design and Development component. The system has proven to work well for
conducting distance education. The project foresees using the AREA-A units in the Sarvodaya District
Centers for conducting distance learning programs that can also be conducted from the HIH or any
location with high-speed internet. Some of the cost can also be absorbed from the users.
To facilitate this technically, a small audio mixer and recording system needs to be installed in the HIH
control room to ensure high quality and consistent recording levels. HIH staff also requires basic
training in audio production techniques including scripting, audio recording and editing, announcing
and on-air presence.
56
12.TECHNICAL ANNEX 1: LIST ACRONYMS AND ABBREVIATIONS
Table 3 List of acronyms and abbreviations used in this report

AC Alternate Current
AREA Addressable Radio for Emergency Alerts
CAD Canadian Dollars
CAP Common Alerting Protocol
CB Cell Broadcasting
CBM Cell Broadcast Message
CDMA Code Division Multiple Access
dB Decibels
DB Database
dBm Decibels referenced to one mill watt
DC Direct Current
DEWN Disaster and Emergency Warning Networks
DEWNS Disaster and Emergency Warning Network Software
DRC Disaster Risk Communication
DWRR Disaster Warning Recovery and Response
EOI Event of Interest
ERP Emergency Response Plan
ERP-C Emergency Response Plan Coordinator
FXP Fixed Phone
GSM Global Standard for Mobile Communications
HIH Hazard Information Hub
HIH-M Hazard Information Hub Monitor
HW Hardware
ICT Information Communication Technology
ICT-G Information Communication Technology Guardian
IPAS Internet Public Alerting System
J2ME Java 2 Micro Edition
KASP Knowledge, Attitude, Skills, and Practice
LKR Sri Lanka Rupees
LM-HWS Last-Mile Hazard Warning System
MOP Mobile Phone
57
NEWS National Early Warning System

NEWS:SL National Early Warning System Sri Lanka
PC Personal Computer
PRA Participatory Rural Appraisal
PSTN Public Switch Telephone Network
RAD Remote Alarm Device
SCDMC Sarvodaya Community Disaster Management Center
SMS Short Message Service
SMSC Short Message Service Center
SW Software
TOT Training of Trainers
USD United States Dollars
VSAT Very Small Aperture Terminal
WiFi Wireless Fidelity
58
13.TECHNICAL ANNEX 2: WHAT IS A COMMUNITY-BASED LM-HWS

Figure 17 illustrates the systems of an end-to-end early warning system. The paper noted that although
the issuing of public hazard warnings was the responsibility of the government, it is unlikely that the
Last-Mile of such a system can be provided solely by government. Rather, it requires a partnership of
all concerned including government, private and non-government sectors.
Global Systems National Systems Last-Mile Systems
National Information
Detection and Message
Warning Communication Community
Modeling Relay
Center Technology
Figure 17 System architecture of an end-to-end hazard detection and warning chain of systems
The LM-HWS architecture depicted in Figure 18 complements the traditional public alerting system
design usually established by local and/or national governments. A traditional public alerting system
issues warnings directly to communities via broadcast media such as television and radio, or through
designated public address (PA) systems. By contrast, the LM-HWS project architecture establishes a
closed user group of first responders, who are equipped with addressable wireless devices for receiving
bulletins issued from Sarvodaya’s Hazard Information Hub.
National Early
Warning
Center ERP-C(n,1)
ICT-G (n)
Village (n)
Terminal
Hazard ERP-C(n,2)
Devices
Information ICT Networks

Domestic & Hub
International
Sources
ERP-C(n,m)
Figure 18 Community-based Alert and Notification communication architecture
For testing purposes the LM-HWS was to design to replicate the components of the typical NEWS; i.e.
a system equivalent to an end-to-end early warning system excluding the “detection and modeling
system”, illustrated in Figure 17. The reader can map the components in Figure 18 to the subsystems in
Figure 17 using the color scheme used to exemplify the identities. For research purposes the closed
user group based system illustrated in Figure 2 will provide the identical functions of the NEWS,
which is an aggregate of the National and Last-Mile Systems. Hence, the model in Figure 2 is an ideal
59
replica to test the functional performance of envisaged NEWS.
A simplified information flow for the LM-HWS is as follows: staff members (HIH-Monitors) at the
HIH monitor hazard events around-the-clock using the Internet. When a potential threat is detected by
an HIH-Monitor (HIH-M), the HIH activates its Emergency Response Plan (ERP) by issuing a
message to the n -number of communities at risk using a combination of wireless ICTs to reach local
first responders (denoted by the arrow between the HIH and ICT-G blocks in Figure 10). Each
community has assigned a person or persons to be responsible for managing the wireless device and
monitoring it for incoming warning messages. This person has received training from Sarvodaya and
is designated as a community ICT-Guardian (ICT-G). When the ICT-G receives a warning message at
the HIH, they are responsible for activating the community-level ERP. The community response will
vary depending on the content of the message, including its priority level. During activation, the ICT-
G informs the m -number of ERP Coordinators (ERP-C), consisting of a First-Aid team, Evacuation
team, Security team, and Message Dissemination team. The Message Dissemination team then relays
the message village-wide through various methods, including as word-of-mouth, ringing local temple
bells, loudspeaker, and so forth.
Message content is encoded using Common Alerting Protocol (CAP), an open source data interchange
standard that includes numerous fields intended to provide consistent and complete messages across
different technologies. The implementation of CAP in the LM-HWS is an important aspect of the
project because it is key in establishing an “all-media” warning capability. Section V describes the
implementation strategy of CAP in Sri Lanka.
14.TECHNICAL ANNEX 3: INTERNETWORKING PIPES, AND GATEWAYS
Figure 19 Microwave Internet link Figure 20 VSAT Internet Link
Initiating point of internetworking is the HIH. The message composed by the authorized user: HIH
Monitor is delivered by Internet Browser based applications or telephony. There are two internet
access paths: 203.888.69.241 Speedcast VSAT, Figure 20, or 202.69.197.113 Dialog Microwave links,
Figure 19. The VSAT 2.4m satellite sits 2m above ground in the front of the SCDMC building and the
Microwave antenna erected 7m above ground at the back of the SCDMC building. VSAT faces east
60
towards the AsiaSat II satellite and the Microwave antenna faces North towards the closest base station
in Kalubowila. Both links are connected to the 16 port CISCO Layer 3 switch sitting inside the HIH
and they are patched such that the HIH PCs operate through the VSAT link and the rest of the users in
the SDCMC building use the Microwave link. In the event the VSAT link goes down then the
microwave link can be patched to HIH PCs. VSAT provides 512Kbps capacity and the Microwave link
offers a 228Kbps capacity. The VSAT hopes from the HIH avoiding any other communications
infrastructure in Sri Lank to Hong Kong. The VSAT has been proven to work on UPS power when
main grid power had been halted. Figure 21 illustrates the primary internet access gateways; i.e. the
yellow boxes.
Colombo CDMA Tower

PSTN
CDMA Tower CDMA Phones
Server
Sri Lanka Telecom Group
Singapore
203.88.69.241 AsiaSat II
Ottawa
64.26.169.57 TCP/IP PCs
AsiaSat-II
Server Solana Group
HIH
203.88.69.241
Hong Kong Colombo
203.88.69.241 202.69.192.51
Server
GSM Tower
Monitor Speedcast Server
HIH Colombo
GSM Devices
202.69.197.113 172.40.1.249
Server
Dialog Group
Melbourne
203.4.254.115
AsiaStar
Touluse
82.225.29.106
Server
UDP/IP Recievers
Singapore
58.185.127.202
Admin WorldSpace Group
Figure 21 ICT providers, gateways, and technologies linking HIH with the Last-Mile
Through the Web based applications: ANNY, DEWN, and IPAS the HIH can issue text messages to
the WorldSpace, Dialog, and Solana Networks gateways. The HIH also has Landline, CDMA and
GSM telephony for voice communication or issuing of fax. In the event the Internet fails then the HIH
will use the telephony to reach the overseas hubs such as the WorldSpace stations in India or
61
Singapore, Solana Networks in Canada to help issue the messages. The internet based IPAS would still
fail if Internet in the entire country is ill-fated. However, the satellite based WorldSpace receivers will
still be functional. Same applies to the Dialog system, which can be accessed by an overseas of
authorized party with access to the internet and the application can issue a message, which can be
received on the RAD and Mobile Phones as they both work on the SMS platform.
As illustrated in Figure 21 the WorldSpace system has several redundant paths and gateways to uplink
the alert messages to the AsiaStar satellite. Once released from the satellite, as long as the terminal
devices on ground are operational, the messages will be received within 7 seconds. The Dialog system
uses their terrestrial GSM network to deliver the alert messages to each of the targeted RAD and
Mobile hand held. Solana Networks uses the TCP/IP internet to send and receive alert messages. The
CDMA telephony system uses the PSTN and the CDMA terrestrial network to communicate via voice.
15.TECHNICAL ANNEX 4: RESEARCH MATRIX AND ICT ALLOCATIONS

Table 4 Village categorization relative to the training regime and organizational ranking
Quadrant Village characteristics
Upper Left - Given ERP Training and they are Less Organized (level 1,2,3) Communities
Upper
- Not given ERP Training and they are Less Organized (level 1,2,3) Communities
Right
Lower Left - Given ERP Training and they are Organized (level 4) communities
Lower
- Not given ERP Training and they are Organized (level 4) communities
Right
Table 5 The deployment of ICT combinations in the 32 selected Sarvodaya communities

AREA
FXP MOP FXP
VSAT MOP Moratuwe MOP AREA
Thirukadalar Meddhawat Oluville
Urawatha Nidavur lla Thamiluvil Maggona
(Trincomale ha (Kalmun
(Galle) (Batticaloa) (Colombo (Kalmunai) (Kalutara)
e) (Matara) ai)
)
AREA +
AREA + Control AREA + AREA +
RAD AREA + AREA + Control
FXP Village RAD FXP
Modara- MOP MOP Village
Wathegama Abeyasin Thondama Shithani-
pallassa Palmunnai Munnai Modara
North ghe Pura nar kudipuram
(Hambanto (Batticalo) (Jaffna) (Colombo)
(Matara) (Ampara) (Jaffna) (Kalmunai)
ta)
62
VSAT MOP MOP

AREA FXP AREA
Modaraga MOP FXP Satur- Samodhaga
Panama Indivinn Brahaman
ma Diyalagoda Periyakallar kondagnya ma
North a a-wattha
(Hambanto (Kalutara) (Batticalo) (Batticallao (Hambanto
(Ampara) (Galle) (Galle)
ta) ) ta)
AREA +
AREA + Control AREA +
FXP AREA + AREA + AREA + Control
RAD Village MOP
Samudraga MOP RAD FXP Village
Kalmunai Mirissa Thallala
ma Valhengoda Venamulla Kottegoda Thalpitiya
II South South
(Trincomale (Galle) (Galle) (Matara) (Kalutara)
(Kalmunai) (Matara) (Matara)
e)
Communities (District defined in parenthesis) that participated in the Last-Mile HazInfo Project. Table
is subdivided in to 4 quadrants with 8 villages in each. VSAT – very small aperture terminal, MOP –
mobile phone, FXP – fixed phone, AREA – Addressable radio for emergency alerts, RAD – remote
alarm device, and Control Village – community without an ICT. The other literals that are different
from the acronyms: VSAT, MOP, FXP, AREA, and RAD along with the Control Villages in Table 5 are
the names of the community; where the literal in the “( )“ is the name of the District the community
belongs to. A combination of the ICTs were tested among 28 ICT-G (32 communities minus 4 control
villages). This allows the project to scrutinize the most user friendly combination of ICT for the ICT-G
in Sri Lanka; where the performance of the HIH-M and ICT-G is not restricted to 1 ICT.
Table 6 Synopsis of Sarvodaya development stages

Develop-
Description of organizational (development) level
ment Stage
Stage 1 Inquiry from the village and organization of an introductory shramadana camp for the
village, during which problems are analyzed together and needs identified.
Stage 2 Establishment of various groups (children’s, youngsters, mothers’ and farmers’
groups), construction of a child development center, and training of staff.
Stage 3 Program for meeting the basic needs and setting up institutions (including the found-
ing of the Sarvodaya Shramadana Society, which is responsible for the village’s de-
velopment initiatives);
Stage 4 Measures to produce income and employment; establishment of complete self-re-
liance and self-financing;
Stage 5 Support for other village communities.
63
16.TECHNICAL ANNEX 5: CALCULATING THE EFFICIENCY OF PROCESSES
Let ti be the time of process i = 1,2, is actually initiated; where E ( ti ) is the expected time at which the
process should have been initiated and ti' is the time process i is terminated and E ( ti′ ) is the time the
process is expected to be terminated. Further introduce a varaible indicated in equations (1) and (2):
Ti = t i′ − t i (1)
is the time interval taken to complete process i ; where
E ( Ti ) = E ( t i′ − t i ) (2).
Figure 22 illustrates the three main time intervals T0 , T1, T2 evaluated in this project. The difference of
the hazard initiating time and the hazard impact time is defined as the period T , which varies for the
combination of each hazard-category and event. Usually the event information will inform the hazard
travel time. A good message will contain the event information such as the vector (speed and direction)
of hazard as well as the time and location of the event. Therefore, the calculation of the minimal
allowable period is obtained by appying equation (3):
T= d (3)
s
where d is the minimum distance between epicenter and impact area and s is the speed at which the
hazard is travelling.
Reliability of the processes are measured as a function of time that it takes to complete the sequence of
processes such as the HIH Monitor relaying alerts to the Last-Mile, Community-First-Responder
relaying alerts to the ERP-Coordinators, and activiating local ERP as illustrated in Figure 18.
Efficiency is calculated by applying the equation (4):
 T − E ( Ti ) 
Ri = 1 −  i  (4)
 T 
when Ti > E ( Ti ) ; otherwise Ri = 1 because the process was completed prior to the anticipated time.
Similarly, the other extreme is for two consecutive processes (example: i = 1 < j = 2 ) from the set of
processs; when ti > t j , then Ri = 0 because the process has surpassed the maximum allowable time for
that process to cpmplete.
Figure 22 illustrates the sequence of functions invloved with issuing an alert message to the last-mile
communities. The had segregated the entire communication sequence of functions in to three
processes: HIH Monitor, ICT, and Community. As labeled in Figure 22, the time periods T0 , T1 , T2
correspond to the three procceses. The time stamps were recorded at several intervals of the sequence
to estimate the efficeincy of each of the processes.
64
HIH-Monitor
t0
Download Alert()
Activate HIH ERP()
T0
ICT Guardian
Relay Alert
t1
Resolve CAP Alert()
Activate ICTG ERP()

Acknowledge
T1
ERP Coordinators
Dissemminate
t2
T2 Activate Community ERP()
Report Status
Relay Results
Figure 22 Sequence diagram of the actors and associated functions of LM-HWS
65
17.TECHNICAL ANNEX 6: MEASURING EFFECTIVENESS OF THE HIH AND

COMMUNITY
Effectiveness of the HIH and Community is a measure of the HIH-M and ICT-G capability to encode
and decode a CAP message. In the case of the HIH-M they should be capable of transforming any
bulletin received from an external source to fit the form of a CAP message. This task is done with the
aid of a form called the “EOI” (Event of Interest) shown in Appendix J. The EOI form contains all the
attributes of a CAP message and is concarted in a way to assist the HIH-M with the task of encoding a
CAP message. The form has a set of check boxes for each of the CAP “qualifier” elements (i.e. header
of the <alert> segment, see Figure 41). By simply ticking the correct check boxes that correspond to
the set of predefined values, the HIH-M can determine the appropriate values to pupulate the qualifier
elements. The difficulty is in populating the <info> segment for each of the languages in the CAP
Profile (see Figure 41). The <info> segment cannot be neglected because it carries the important
elements that describe the priority of the alert message. Although the CAP ontology allows the <info>
section to be optional the CAP Profile for Sri Lanka makes it a mandetory element. HIH-M will extract
this information from the bulletines to the best of their ability. Never the less, in all cases the entire
payload of the bulletin received at the HIH will be implanted in the <description> element. Apply
Table 9 in section 20.2 to obtain a real valued score for the HIH-M’s effectiveness to complete the CAP
message with the essential elements.
The ICT-G must be capable of interpreting the CAP message received from the HIH and most
importantly be able to determine the priority of the message, which is determined by reading the
<urgency>, <sevierity>, and <certainity> elements embeded in the CAP <info> segment. In the
absence of the <urgency>, <sevierity>, and <certainty> elements, the ICT-G must be capable of
identifying the information for these elements from the <description> element of the CAP message.
ICT-G were provided with a paper form: Alert Log (Datasheet 1 in Appendix J), which contained the
essential elements needed for documenting the CAP message. The same Alert Log would provide them
with the tools to extract the priority and description of the alert message. Once again by applying the
methodology described in section 20.2, a real valued measure for the ICT-G effectiveness can be
obtained from Table 9.
ERP-C are expected to activate all their ERPs with respect to the Hazadous event invloving the
activation of the local dissemination teams, first-aid teams, security teams, and evacuation teams. They
are further expected to clearly and accurately deliver the information related to the incident as well as
cover all the households in their juresdiction.
18.TECHNICAL ANNEX 7: CALCULATING RELIABILITY OF ICT

The basic question governing the reliability measure is “did the ICT based system work on the day of
the live-exercise?” Reliability, denoted by R can be measured in at least two aspects: certainty and
efficiency; denoted by Rc and Re respectively. Whereas certainty refers to the operational state of a
device on the day of the exercise, efficiency measures the time taken to complete the transmission of a
message in relation to the anticipated hazard risk (i.e., will the message be received with enough
66
advance warning to take action?).
Applying equations (4) for efficiency to the ICT segment of the sequence and equation (6) for the
certainty of ICT, the overall reliability R of the wireless ICT in a LM-HWS is computed as functions
of the certainty and efficiency of the ICTs; as given by equation (5):
R = Rc × Re (5)
18.1.Certainty of Message Receipt
Enumaeration of the Effectiveness Parameter:

Signal Strength in Terminal Devices
y max
1.00
0.90
0.80 Minimum
Effectiveness Score
0.70 to receive
alerts
0.60 y min
0.50
Minimum to
0.40 gain
0.30 service
0.20
0.10
0.00
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
Number of Bars
Figure 23 Graph for calculating the effectiveness based on signal strength
In some situations the ICT failed on the day of the exercise and was given a reliability score of 0 in
terms of certainty. It was evident that the strength of the signal coverage was catalyst to an ICT-G
undoubtedly receiving an alert. Signal coverage measures the certainty of a message reaching an ICT-
G recipient at various times of day and is intended to reflect the variability between mobile, nomadic,
and fixed wireless devices. Signal strength is important for the Alerting feature because if the signal is
weak then some data can get lost. As a result the terminal device may not get the full bit string,
causing failure to activate some of its active alerting function sensory features that are triggered by
designated bits in the data string.
Wireless signal is usually measured as a function of the power of the signal in decibels (dB) then
referenced to 1 mill watt (dBm). The signal strength was measured at the ICT-Gs home or the location
where equipment was installed. A GSM mobile phone can function on -104 to -47 dBm range,
Satellites operates on -127 to -60 dBm range, and CDMA phone operates on -106 to -48 dBm range.
All the devices used in the project display their signal strength using illumination of histogram bars (or
67
LEDs) that are a function relative to the dBm range of the device.
y min y y max
Figure 24 Example of signal strength bars
Given the device has a total of ymax bars, at a particular location such as the ICT-G’s home, the device
may have y ≤ ymax number of bars illuminated; where y is a real number.
EXAMPLE 2: A particular terminal device may have the bars color coded; where a green bar implies full
strength, amber bar implies half strength, and red bar implies zero strength. As illustrated in Figure 24,
a color coding for a device with a maximum of 8 bars having 3 green bars, 1 amber bar, and 4 red bars
illuminated may imply signal strength of y = 3.5 .
For a particular application of the ICT (e.g. application of sending and receiving SMS) to work
properly the service provider can specify a minimum requirement for the signal strength (such as 1
bar). Let us denote this minimum strength by ymin . The device is given a score based on the signal
strength indicated by the bars by applying equation (6):
 1 , y ≥ y min
Rc =  1 + e ymin − y (6)
 0 , y < y min
The exponential function defined in equation (6) and shown Figure 23 will give a score between 0 and
1 for certainty. When y = ymin the score will be 0.5, implying that with the terminal device indicating
minimum signal strength there is only a 50% chance that the alert message will come through without
any doubt. As the number of signal strength bars: y > ymin increase the score will reach a value of 1 at
an exponential rate. Similarly as the number of bars decrease below the minimum number of bars
needed the probability of receiving the alert drops below 50 percent.
The calculation takes in to consideration that ymax − ymin ≈ 5 because all of the ICT terminal devices
tested in this project had either a 5 bar or 7 bar signal strength display; where all of the devices with 5
bars required approximately minimum application specific requirement of 2 bars and the devices with
a 7 bar display required approximately a 3 bar minimum requirement. It is evident that the signal
strength scoring equation (6) has to be refined and generalized for all ICT terminal devices. For the
purpose of this research equation (6) was sufficient. The authors recognize that equation (6) does not
take in to considerations of uncertainties caused due to operator (human) mishaps that are discussed in
section 3.1.
68
18.2.Efficiency Measure of Transmission
Equation (4) estimates the efficiency of the ICTs in terms of a time taken to transmit the message from
the HIH to the ICT-G. The time period begins when the HIH-M starts entering the message in the web
based application or started dialing the phone and the time period ends when the ICT-G guardian
completes recording the message in the Alert Log (Datasheet 1, Appendix J).
19.TECHNICAL ANNEX 8: MEASURING EFFECTIVENESS OF THE ICT

19.1.Effectiveness of the Individual ICTs
Effectiveness was measured as a function of a set of discrete parameters. The project has defined 11
such discrete parameters: language diversity, full CAP capability, audio and text medium availability,
bi-directionality, total cost of ownership, DC power consumption, daily utilization, acknowledgement
of message receipt, active alerting functionality, weight of wireless ICT, and volume of terminal
device. Each parameter is denoted by a literal q1 , q 2 ,  , q11 . A continuous function or “Liken” type
rating (i.e. discrete function) is defined and applied to obtain a real valued score between 0 and 1 for
each literal, which is denoted by P( q i ) for i = 111 . A conjunction of the literals q1 ∧ q 2 ∧  ∧ q11 defines
the design requirement for an effective ICT in a LM-HWS. Hence, a single score for the effectiveness
G of the deployed wireless ICT configuration is obtained by multiplying the real valued score of all
the literals, as stated in equation (7).
G = G ( q1 ) × G ( q 2 ) × × G ( q11 ) (7)
The 11 parameters are further grouped in to 5 cliques: CAP Completeness, Two-way, Adoptability,
Miniaturization, and Alerting; as shown in Table 7, which is not listed in any particular order as their
contribution to effectiveness in a Community-based LM-HWS is equally important.
Table 7 Cliques and the set of corresponding parameters

Clique Parameter description Abbreviation
Adoptability Integration in to daily life or development practices Utilization
Total cost of ownership of ICT services and terminal Affordability
device
Alerting Acknowledgement of message receipt Accountability
Active alerting function Wakeup
CAP Complete Language diversity Ethnicity
Full CAP capability All-media All-hazards
Audio and Text mediums of communication Multimedia
Miniaturization Weight of wireless ICT terminal devices Weight
Dimension of wireless ICT terminal devices Volume
DC power consumption Longevity
Two-way Upstream and Downstream communication Bi-directionality
69
The researchers believe that it is better to scrutinize the effectiveness in terms of discrete cliques:
G1 , G2 , G3 , G4 , G5 opposed a single real value G because in cases such as the AREA which is one-way
broadcast technology would score a 0 in terms of two-way capability and would drive the overall
effectiveness score G = 0 and hide the true capabilities in which the effectiveness of the technology
ranks very high in all other capabilities.
19.2.Effectiveness of a Combination of ICTs
As shown in Table 5 of the research matrix, fourteen communities were equipped with two ICTs.
When measuring the effectiveness for the combination of ICTs (i.e. two or more) the maximum value
of the particular parameter of the set of ICTs is taken as the value for the set. Let j = 1,2,  , m be the set
of ICTs; where G j ( q i ) be the real value effectiveness measure for literal q i . Then
{Max(G j ( qi ) j = 1,2,  , m)} is the effectiveness measure for the combination of ICTs.
20.TECHNICAL ANNEX 9: FUZZIFICATION OF FULL CAP COMPLETENESS

In this project we define a Complete Full-CAP Message to be one that complies with the CAP Profile
for Sri Lanka, contains all three languages: Sinhala, Tamil, and English, and also is disseminated in
modes of Audio (i.e. Voice) and Text. The final rating is the multiplication of the values obtained from
Tables 2, 3, & 4.
20.1.CAP Profile for Sri Lanka
CAP adopts a Document Type Definition (DTD) Extensible Markup Language (XML) data structure
that consist of a main element <Alert> and subelements <Info>, <Area>, and <Resources> as
illustrated in Figure 41 in Technical Annex 15. The <urgency> code denotes the time to impact of the
event. <severity> codes denotes the scale of impact of the event. <certainty> Code denotes the
probability of the event. These 3 elements define the Priority of the <event>. The Priority is a higher
order function that maps Urgency, Severity, and Certainty values to a distinct Priority: Urgent, High,
Medium, or Low. The mapping is discussed in Table 8.
Table 8 Matrix to determine Message Priority with CAP elements

Priority <urgency> <severity> <certainty>
Urgent Immediate Extreme Observed
High Expected Severe Observed
Medium Expected Moderate Observed
Low Expected Unknown Likely
The reader may refer to the “Guidelines for the HIH in Technical Annex 15 to fully comprehend the
CAP Profile for Sri Lanka, which elaborates each of the elements and defines the set of predifined
70
values for the for the elements with fixed set of values.
20.2.Compulsory Elements of the CAP Profile
A CAP message is defined to have a high effectiveness value of 1 if the message contains all CAP
elements described in Tables 20 - 23 in Technical Annex 15. The lower end value 0 is when the
message is an empty CAP message; i.e. dead air or text elements with null values. The compulsory
Elements of the CAP Profile include elements in the <Alert> “qualifier” elements: <Incident>,
<Identifier>, <Sender>, <Sent>, <Status>, <msgType>, <Scope>, and the “sub” elements: <Info>,
<Resource>, and <Area>. Table 9 defines the discrete function which enumerates the capability of a
technology.
Table 9 Scaling function for Full-CAP Capabilities

Value Fuzzy Rules
1.00 All sub elements that are contained in the <Alert> element, which includes all the
qualifier elements and <info> element as well as the <resource> and <area> sub
elements
0.95 Mandatory elements described in the Profile for Sri Lanka, which are qualifier elements
in the <alert> segment and the sub elements of the <Info> element with at least the
<urgency>, <severity>, <certainty>, and <description>
0.85 Elements of the <alert> segment and the <info> sub element <description> only
0.70 <description> only
0.50 Elements <category> and <event> only
0.25 Mandatory sub elements of the <alert> segment only
0 Otherwise
20.3.Language Diversity in Sri Lanka
The function to enumerate the language diversity is obtained by applying basic set theory. The [ n] set of
linguistics recognized and used in a area (community, state, province, country, or region) with a
population set X , the literals Li ( X ) define the subset of the overall population that can speak, read, and
write the language i ; where i = 1,2,  , n . For two languages i, j ∈ {1,2,  , n} : Li ( X ) ∪ L j ( X ) is the subset of
the total population that can speak, read, and write either or both languages. The ideal case is the
technology having the capability to carry alert messages in all the languages covering the entire
population in the targeted area; consequently would score 1. The lower end score of 0 is given to a
technology that would cover none of the languages applicable to the area in which the alert messages
are issued to. Let [ m] ⊆ [ n] be the set of linguistics of the universal set for linguistics applicable to the
area. The real value calculated from equation (8) gives a score between 0 and 1 indicating the fraction
of population “covered” by the subset of language relative to the overall population.
 j∈ [ m ] L j (X)
(8)
X
71
The rules in Table 11 are constructed from equation (8) by applying the Ethnicity Statistics 7 obtained
from the Census Bureau of Sri Lanka; approximately 82% are Sinhalese, 17.3% are Tamil (Sri Lanka
Moor, Sri Lanka Tamil and Indian Tamil), and the rest, 0.7% are Other (Burgher, Malay, Sri Lanka
Chetty, Bharatha, etc). “Other” ethnic groups are literate in English and in a major portion of them can
speak and read either Sinhala or Tamil. The Census Bureau provides data on the ability of an ethnic
group’s literacy in the secondary languages. Table 10 is an extract of the data distribution by ethnic
group, which is used to generate a Venn diagram to obtain the necessary counts for the language sets:
Sinhala, Tamil, and English. Since the percentage of the population who are literate in all 3 languages
is relative very small, an assumption is made that the size of this set is 0.
Table 10 Percentage of population by ethnic group (race) literacy in the second languages
Race  Sinhala Tamil Other
Literate in  Tamil English Sinhala English Sinhala Tamil
Percentage  3.00 9.83 15.00 5.00 0 0
The percentage3 calculated in Table 10 is with respect to the overall population and not relative to the
race. The statistics relative to each race is given in the data sheet provided by the census bureau. The
notation in the column fuzzy rule: Sinhala + Tamil + English imply the set union of the populations
speaking Sinhala, Tamil, or English.
Table 11 Scaling Function for Language Diversity

Value Fuzzy Rule
1.00 Sinhala + Tamil + English
0.99 Sinhala + Tamil
0.95 Sinhala + English
0.80 Sinhala Only
0.38 Tamil + English
0.28 Tamil Only
0.15 English Only
0 Otherwise
20.4.Mix of Pictorial, Audio and Text Communication Mediums
The project used Table 12 to weight the ICT as a function of the capability to disseminate pictorial,
audio or text messages. The context of pictorial refers to a message depicted by a photograph, digital
image, or non alphanumeric symbol that can be interpreted clearly and accurately as a warning
message. The message must be able to provide the necessary and sufficient set of risk information or
guide towards an alternate source to fetch the full set of information. The picture may contain text in
one or more languages to describe the scenario. A simple example is a Satellite image of cyclone on a
map with the direction of movement indicated by an arrow and the category indicated by a color. An
audible message such as what one would receive from a radio broadcast, public announcing system, or
7
Statistics used in the explanation was obtained from -- http://www.statistics.gov.lk/census2001/population/district/t001c.htm ; the
values used for Rural and Urban as a collective.
72
television news cast are considered as audio messages. The medium of text is the use of alpha numeric
characters (numbers and language alphabet) to communicate the alert message. Computers use a
standard called Unicode to define characters of a particular language where the Unicode maps to a
hexadecimal value that can be interpreted by any electronic processor.
Television is a terminal device that is capable of simultaneously communicating via all three mediums.
Digital radios such as the AREA device used in this project can deliver both audio and text. The SMS
based alerting technique via mobile phone is only capable of text messaging.
Table 12 Scaling Function for Audio and Text Communication Mediums

Value Fuzzy Rule
1.00 Pictorial + Audio + Text
0.95 Pictorial + Audio
0.90 Audio + Text
0.80 Audio
0.70 Pictorial + Text
0.60 Pictorial
0.50 Text only
0 Otherwise
21.TECHNICAL ANNEX 10: FUZZIFICATION OF ALERTING PROCESS

The project defines alerting to be a function where a pictorial (image), written (text) or voice (audio)
message is relayed to a Community-First-Responder (person manning the ICT device); in this project it
would be via an ICT device belonging to the Community-First-Responder. Simple issuing the message
alone is inadequate. Therefore, the Alerting function is complete if the message is received by the
Community-First-Responder and the Community-First-Responder returns an acknowledgement
message via same or alternate path to the Sender at the Message-Relay.
The sequence of functions are 1) Hub Relays a message in CAP format to the ICT Providers 2) ICT
Provider resolves message and transmits to the respective devices 3) devices receive alert 4) device
activates wakeup function 5) Community-First-Responder acknowledges message by turning off
wakeup function 6) device auto-transmits or Community-First-Responder sends acknowledgement to
ICT Provider’s database 7) on demand Acknowledgement Report made available for Message Sender.
Figure 22 in Technical Annex 4 illustrate the sequence of functions and the actors: HIH Monitors and
ICT Guardians associated with carrying out the function.
As described by the cliques in Table 7 in Technical Annex 8 the project has defined the following
functional indicators: Message Receipt Acknowledgement (accountability) and active alerting
(wakeup) to assess the effectiveness of the ICT as an Alerting tool.
73
21.1.Acknowledgement of Message Receipt
Message receipt acknowledgement is vital for the message Community-First-Responders to inform the
Sender that the message was received. In the context of Alerting the acknowledgement must be
initiated by a human. Hence, the devices must have a method in place to send an acknowledgement as
soon as the Community-First-Responder accepts the message. The Message Receipt an
Acknowledgement is a Probability function anchored on the time taken between Alerting functions 1)
and 7). The device would score 1.0 if the acknowledgement was reported in less than 1.0 minutes;
score 0.95 if it took less than 5.0minutes; score 0.85 if it took less than 10.0 minutes; score 0.70 if it
took less than 20.0 minutes; score 0.50 if it took less than 45.0 minutes; score 0.25 if it took less than
90.0 minutes. The probability density function defines a sigmoid shaped discrete scaling function;
where the score is higher if the time taken to report acknowledgement was al most instantaneous.
Table 13 Scaling Function for Acknowledgement of Message Receipt

Value Fuzzy Rules
1.00 < 1.0 min
0.95 < 5.0 min
0.85 < 10.0 min
0.70 < 20.0 min
0.50 < 45.0 min
0.25 < 90.0 min
0 Otherwise
21.2.Wakeup function to gain attention
Wakeup function is designed to draw the attention of the targeted Community-First-Responders with a
combination of attention-getters: sounding-sirens, flashing-lights, or mechanical-vibrations. Basically
the Wakeup feature should actuate the aural, visual, and sensual sensors of the Community-First-
Responder. The ICT device scores 1.0 if it has all three attention getters; scores 0.95 if it has a Siren
and Light because this combination does not require the device to be attached to the Community-First-
Responder in any way; scores 0.85 if it sounds a siren only; scores 0.70 if it sounds a siren and
vibrates; scores 0.50 is it activates a flashing-light and vibrates; scores 0.25 activates only one a
flashing-light or vibrates; scores 0 otherwise. For a person fast a sleep simply a flashing light alone
will not get the attention of the sleeping person and needs an “Alarm” sound to wake them up. When
defining the wakeup function for people with disabilities then the manufacturer must use the best
attention getter to suit the requirement. Figure 26 describes the function for enumerating the ICT’s
effectiveness to wakeup the ICT-G based on the visual, audio, and tactile sensory.
74
Enumeration of the Effectiveness

parameter: Active Alert Function in
Terminal Devices
1.00
0.95
0.85
Effectiveness Score
0.70
0.50
0.25
0.00
l ti le
s e c t il e c t li e n l y c t il e s u a ta c
i ta ta o ta v i &
th e rw l o r l & i b le le & le & u al ,
o a a d i s
v is
u
v is
u au u d i b di b , v
a au ab le
di
au
Active Alert Functions
Figure 25 Enumeration function for wake up function
22.TECHNICAL ANNEX 11: FUZZIFICATION OF TWO-WAY COMMUNICATION

22.1.Bi-Directionality for Upstream and Downstream Communication
Bi-Directionality is an indicator to measure the ability of the device to permit upstream

communications from local communities to the Message Relay as well as downstream communication
from the Message Relay to the Communities. The upstream communication is mainly for Last-Mile
Communities to inquire-of and report situations affecting their communities. The MOP, FXP, and
VSAT allow both upstream and downstream communication without any restriction. The RAD has
limited the upstream communication such that the user can “call back” only when an alert is received.
The AREA does not allow any upstream communication at all. Once again each of the devices is given
a score on a scale of 0 to 1 such that if it has no restrictions then it scores a 1.0 and a lower score for all
other combinations giving prominence to (i.e. higher score) for downstream communication over
upstream communication in the domain of community-based alert and notification. Table 14 provides a
discrete enumeration method to give a real valued score to an ICT based on the upstream and
downstream communication capabilities.
75
Table 14 Scaling Function for Bi-directionality

Value Rules
1.00 Upstream and Downstream No restrictions
0.95 Downstream and restricted Upstream
0.85 Downstream only
0.70 Restricted Downstream and Upstream
0.50 Upstream only
0.25 Restricted Downstream and Restricted Upstream
0 Otherwise
23.TECHNICAL ANNEX 12: FUZZIFICATION OF ADOPTABILITY

23.1.Integration of ICT in to Community daily life
Integration in to daily life is basically a utilization measure. Utilization can be quantitatively measured
in terms of time spent or monthly cost or both. However, there are non holistic measures that can give
greater value to the ICT being adopted such as the WorldSpace radio channel that can be used for
community development, which may have a greater impact to the entire community opposed to a
mobile phone also adoptable for community development but on an individual basis. Similarly the
VSAT that gives access to the internet, email, and VoIP cannot be measured on time alone because the
utility is based on several applications that cannot be ranked on importance.
To measure the utilization the Communities were given a daily journal to record the usage. The main
variable governing the assessment is time spent by the user and the purpose of use as to how it may be
beneficial.
23.2.Total Cost of Ownership
There is a hardware and service cost associated with owning and operating a personal communications
technology. The hardware and service costs are further partitioned as fixed cost and variable cost,
which gives rise to four parameters as described in Table 15. Although the hardware can guarantee
usage for five years, it is standard business practice to linearly depreciate electronic hardware over
three years. Therefore, the total cost of ownership is anchored on the hardware and service costs over a
three year period. In the Community-base hazard information dissemination model a single ICT
deployment, comprising one or two terminal devices, is to cover the entire community. Therefore, the
TCO is uniformly distributed amongst the number of households; i.e. TCO is divided by the number of
house holds.
76
Table 15 Parameters associated with calculating TOC

Fixed cost Variable cost (per month)
Purchasing the terminal device: H fc ; Warranty, maintenance, licensing fees: H vc ;
Hardware brand new or second hand cost of the i.e. usually there is some kind of a recurring
ICT cost, let say the owner decided to insure the
hand set pay a premium on a monthly basis
Entry fees or setup fees: fc ; person has Periodic subscription or rental charges: S vc ;
S
to purchase a sim card, assume they assuming it is a pre paid, the person has to
Service throw away the sim when they throw top it up on a monthly basis
away the phone in 3 years
In the attempt to devise a method to calculate the TCO the following assumptions are taken in to
consideration:
1) “Community-based” early warning systems are for developing nations where governments in
these nations neither have the capacity nor have the motive to deliver the public good to their
citizens; therefore, the community takes the initiative to link with an alternate source through
ICTs
2) More than 50% of the population in developing nations are in socio economic classes D & E;
therefore, the solutions proposed must be affordable to them
3) Calculating the TCO based on the data available for D & E provides a good lower bound for
the average TCO for an ICT in a community-based early warning system for that nation
4) The minimum number of households per community is N = 100
5) Equation (9) is an appropriate formula to calculate the TCO for a community for a period of 3
year
Per household three year TCO =
( H fc + S fc + ( H vc + S vc ) × 36 ) (9)
N
EXAMPLE 3: The project extracted the values for the fixed and variable costs for the hardware and
service variables from the paper “Teleuse on a Shoestring”8; H fc = 95 , H vc = 0 , S fc = 15 , and S vc = 7.23 .
The information is based on year 2006 survey data. The average monthly expenditure S vc for Sri Lanka
is calculated as an average of pre-paid mobile owners, post-paid mobile owners, and home fixed
phones. From the figures in the survey we can obtain that the price of a brand new unit is USD 95. One
time entry fee to obtain a service is USD10. For the real data and applying equation (9) we obtain the
TCO to be USD 3.70 per community house hold over a period of three years. The TCO lower bound is
obtained from information related telephony, which may seem unbiased when using the TCO value to
weigh ICTs such as the VSAT or AREA.
Let us assume the benchmark TCO per community to be E ( C ) ; where E ( C ) = 3.70 as calculated in
Example 3. Let C denote the TCO of the particular ICT. Equation (9) defines a penalty function,
which describes a quantified effectiveness score for the TCO:
8
H. de Silva, A. Zainudeen, (January 2007), “Teleuse on a Shoestring: Beyond Universal Access”, CPRsouth, can be found
here -- http://www.cprsouth.org/index.html/files/Session%205,%20Paper%201,%20January%2019,%202007%20-
%20Ayesha%20Zainudeen.pdf
77
 1 C ≤ E (C )

G( C ) =  C − E ( C ) (10)
C > E( C )
 E ( C )
The graph in Figure 27 describes the basic shape of the function defined by equation (9) for the Sri
Lankan case.
Total Cost of Ownership vs Effectiveness score
1.00
0.80
Effectiveness
0.60
0.40
0.20
0.00
1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00
TCO (USD)
Figure 26 Graphs of TCO penalty function
24.TECHNICAL ANNEX 13: FUZZIFICATION OF MINIATURERIZATION

The Community-based early warning equipment are not restricted to alert and notification but are
expected to be a valuable communication source throughout the entire disaster management cycle of
mitigation, relief, and recovery. Relief workers are often called upon to perform their duties in remote
areas which are not served by telecommunications or where networks have been destroyed.
Coordinating aid efforts and keeping in touch with fellow workers, informing central authorities the
state of affairs of the community are vital to the rapid delivery of assistance where it is most needed.
Hence, the project belief is that the terminal devices used for alert and notification must be easily
portable to strategic locations and be rapidly deployable. The trend in the world is to build electronics
as small and practical as possible. The same notion is envisaged in terminal devices designed for
communication in disaster management. As a result the project has identified weight, volume, and
longevity as parameters that measure what is termed as miniaturization. The key factor of
transportability is a function of the parameters weight and volume of the communications device.
Where as rapid deployment (free of infrastructure) and uninterrupted service becomes a function of the
longevity (available power).
78
24.1.Weight of the entire functional unit
Effectiveness vs weight
1.00
Effectiveness
0.50
0.00
20 30 40 50 60
weight (Kg)
Figure 27 Graph of the equation (11) for weights above 20Kg
The weight must be below what an average adult human being is capable of carrying. The project sets
a standard for the load a human being, without special skills, can carry to be 20 kilograms. If the
terminal device is over 20Kg, then the effectiveness of the device in terms of weight is penalized and
given a score between 1 and 0. If the device weighs less than or equal to 20Kg then it will receive a
full score of 1. A weight of 30Kg is bearable and a single person can carry. However, when the weight
is beyond 40Kg it becomes intolerable for a single person to carry. The project devised a penalty
function that exponentially decreases the effectiveness score as the weight increases beyond 20Kg.
 1 W ≤ 20
G (W ) =  1 ( − W + 20 ) (11)
 e 20 W > 20
The penalty function, described by equation (11), is devised in a way to reduce the effectiveness
exponentially because the intention is to minimize the number of resources allocated to the tasks such
as transporting and operationalizing the communications during a crisis situation When the weight is
30Kg the effectiveness is 0.6 above 0.5 and as the weight is beyond 35Kg the effectiveness score
depreciates below 0.5 and approaches 0 rapidly there on. The reciprocal of the effectiveness score can
be interpret as the number of people required to carry the equipment; i.e. when the weight is over
35Kg, 1G( 35) = 2.12 ≈ 2 , at least 2 people are need to carry the equipment.
24.2.Volume of terminal device and peripherals
The Volume parameter will be associated with the mathematical context of a “packing”; where the
object with a specific volume must meet certain criteria and as a result would score a high or low
probabilistic value. Volume dimensions of the entire device, including all parts necessary for it to be
fully functional, must be such that the device can be placed in a space without extra long pieces
79
protruding making it difficult to transport or place in an indoor/outdoor environment. For example,

there are standard set of dimensions that go to make up a total volume of 114cm where length, width,
and height, approximately, is 56 x 35 x 23 cm, respectively, for airline carryon luggage. Similarly, a
constraint is set that the dimensions terminal device with shapes like a prism, a cuboid, a sphere, or
ellipse, must be less than or equal to 0.5 meters. The value 0.5 is chosen such that the object (terminal
device) can easily fit in a suitcase or backpack that can be easily carried by hand or on the shoulder in a
back pack.
Effectiveness vs Volume
1
Effectiveness
0.5
0
0.125 0.25 0.375 0.5 0.625 0.75 0.875 1
Volume (cubic meters)
Figure 28 Graph of equation (12) for volume above 0.125m3
Since it is inappropriate to compare two objects, the diameter of a sphere against the diagonal length of
a cuboid, for example, to judge a winner, the project has chosen to compare the volume given that that
any dimensions contributing to the volume must be less than or equal to 0.5 meters. Thus the
maximum volume of a spherical object will be approximately 0.06m3 or the maximum volume of a
cuboid will be 0.125m3. From geometry or packing theory it is known that amongst three dimensional
objects the cube with length l has the largest volume of all the geometries such as a prism, sphere, or
ellipse with same dimension of maximum size l . Similar to the weigh penalty function described by
equation (11), the linear penalty function (12) is devised for the volume:
 1 V ≤ 0.125
G (V ) =  2( − V + 0.125) (12)
e V > 0.125
Any thing in between 0.125 and 0.250 will get a score between 0 and 1 depicting the effectiveness in
terms of volume. The same argument as in the case of the weight applies here; where 1G (V ) implies the
number of individual maximum volume containers: suitcase or backpack required to haul the
equipment; i.e. 1G( 0.5) = 2.12 > 2 requires at least 2 containers.
80
24.3.Longevity of operational state
Longevity is included as part of the set measuring miniaturization because the operational life span is
dependent upon the rate of power consumption and capacity of the power source of the wireless
technologies. All the electronics of the communications devices work on Direct Current (DC) power.
The duration of the DC power source is relative to the capacity of the battery powering the device and
the rate at which the power is consumed by the circuitry. All ICTs except the VSAT, tested in the
project, were equipped with rechargeable batteries, which could be charged via Alternate Current (AC)
from the main power grid or via solar power. The 20 – 40 watt VSAT modems can be powered by high
capacity batteries. A device powered by AC can be regarded as having an infinite longevity
(immortal), assuming the service of power is uninterrupted. AC is converted to DC before electricity
flows through the electronics. When main grid AC power is interrupted then the device will switch to
the charged battery power.
Manufacturers are compelled by IEEE standards to specify the input voltage, acceptable range of the
current, and the power consumed by the apparatus. From the theory of electricity it is known that
power is equal to the voltage multiplied by the current. Moreover, current is the rate at which a charge
flows though a point in a electric circuit. Given the capacity of the battery, the utilization time of the
battery (i.e. battery life) can be calculated.
The project set the standard that each wireless device must function for a minimum of 16 hours on
battery. Assuming the battery can be charged using solar during daylight and the minimum
requirement for the strength of sunlight is available for 8 hours, the unit must function for the
remaining 16 hours on battery over the 24 hour period.
Effectiveness vs Longevity
1
Effectiveness
0.5
0
0 4 8 12 16
Longevity (hours)
Figure 29 Graph of equation (13) from 0 to 16 hours
Let L be the battery life and G ( L ) be the effectiveness measure for operational longevity of the ICT;
where G ( L ) = 0 if the battery has 0 capacity; i.e. 16 hours and G ( L ) = 1 if the battery life is over 16
hours. The linearly depreciating penalty function is given by equation (13):
81
 1 L ≥ 16

G( L) =   L  L < 16 (13)
 
  16 
Equation (9) is linear because the battery power is assumed to drain uniformly. As shown in Figure 7
the reciprocal of the effectiveness 1G( L ) can be interpreted as the number of batteries required to
sustain the ICT for 16 hours. A battery with a 4 hour life span: 1G( 4) = 4 implies that 4 batteries are
required to keep the ICT operational for 16 hours.
25.TECHNICAL ANNEX 14: EXPECTED PERFORMANCE OF THE ICT

25.1.WorldSpace Addressable Satellite Radio System
Transmission Technology
There are two types of satellite transponders onboard the WS satellites, one transparent and the other
on-board processed. The Transparent Repeater repeats Time Division Multiplexed (TDM) carriers sent
to the satellite from large earth stations at X-band onto downlink carriers at L-band. Each satellite has
three spot beams with each beam containing two TDM carriers centered at different frequencies in the
band 1452-1492 MHz. Each beam covers a large geographical area and with the six beams on the two
satellites in service, the coverage extends to more than 130 countries in Asia, Africa and Europe.
Each TDM carries 1.536 Mbps, divided into 96 Prime Rate Channels (PRC) each of 16kbps. A basic
audio program can use just one PRC and achieve an audio quality better than analog short wave as WS
uses efficient audio coding scheme MPEG Layer-3. As many as 8 PRCs can form one Broadcast
Channel (BC) depending on the desired quality. The WS digital format incorporates interleaving, 255,
223 Reed-Solomon block coder and Rate ½ Vitterbi convolution encoding technologies to protect the
service against transmission errors. At any given geographical location, typically 30-40 subscribable
audio channels are available, giving the listener enough choice.
AREA Terminal Device Components
The receiver, called DAMB-R2, is a low-memory radio with a small display and a limited processing
power. DAMB-R2 features dual channel reception with one of the channels designated as a data
channel called OAAC (Over Air Activation Channel). Through the OAAC channel the alert provider
can send an alert message. DAMB-R2 has the capability to monitor the alert, validate the message and
perform the specified action. The action could range from activating a relay for a siren, turning
on/switch to a WS channel for audio messages and displaying text regarding the alert.
An external box can be fitted via a USB to the DAMB-R2 to enhance the display with a larger digital
display as well as couple other peripherals such as an audible siren and GPS receiver. A palm-sized
linear patch antenna was used to receive the services. All WS specific processing including MPEG
82
decoding is handled by the StarMan chipset, which is incorporated in all the receivers. Each radio
receiver features a BC digital output connector allowing to access to the full content of the selected BC
from an external appliance. Figure 30 illustrates all the components associated with the AREA system.
Figure 30 Components of the AREA Terminal device
States of the AREA Receiver
Power On Tune Channel Cancel Alert
Power Off Play Audio Sound Siren
Recieve Alert Display Tex Play EW Audio
Figure 31 State transition diagram for the AREA systems
After power is supplied, AREA (B,C & M) can be initiated from the Power Off state to the Power On.
At this point any of the available channels can be tuned and it starts Playing Audio. The AREA
receiver is now ready to receive alerts. Once an alert is received, the AREA normally Sound Siren and
Display Text of the mandatory elements of the CAP message. In the case the channel ID (BCID) is
specified in the CAP message, then DAMB-R2 can be automatically preset to that channel and this can
be chosen as the Emergency Warning (EW) broadcast channel. Once the alert cancellation is received,
it stops displaying the text and the tuner once again becomes active for selection of the available
channels. Figure 31 shows the various operational states of the AREA system.
83
ANNY Network Interface
The uplink segment includes a special, secure, web interface, hosted at one or more of the designated
centers from where the Emergency Warning message is generated; see Figure 14. Additionally, the in-
terface incorporates procedures for authentication and logging of the alerts, generation, uploading and
scheduling of transmission of content for the supplementary audio channel that WS supports once an
alert is announced, updating an announced alert, canceling the alert, periodic end-to-end testing of the
entire system.
25.2.Dialog Disaster and Emergency Warning System
Figure 32 High level architecture of the DEWNS with network and Terminal Devices
The Disaster Emergency Warning Network (DEWN) is based entirely on widely available mobile
communications technologies such as Short Messages (SMS) and Cell Broadcast Messages (CBM),
aimed at rendering a cost effective and reliable mass alert system. The system is compliant with CAP.
As shown in Figure 32, DEWN comprises of two basic elements – i.e. the DEWN Server and DEWN
Clients. The DEWN Server will reside in a secure facility and will be used by authorized persons to
generate warning messages via SMS or CBM. The DEWN Clients are the intended recipients of the
above mentioned messages. Upon reception of the messages the clients will take necessary measures to
inform the users of the system about the warning. The DEWN Clients are twofold. They are a
Java/Symbion application for mobile phones developed by Microimage and the Remote Alarm Device
84
discussed developed by University-of-Moratuwa Dialog Mobile Communications Research Lab.

Messaging Software Application
At this stage of the project, MicroImage has developed a web browser based CAP GUI; namely the
Disaster and Emergency Warning Network Software (DEWNS). The GUI interfaces with the GSM
and the Dialog SMS components. The Dialog Telekom SMS servers are registered as teleports in their
respective CAP alerting software. SW application for issuing alert messages is accessed via an Internet
web-based connection. The authorized user logs in and proceeds through a sequential 3-screen process
to create and compose a CAP-compliant message. Presently, the RAD/SMS CAP software, shown in
Figure 15, operates independently of other LM-HWS components.
Remote Alarm Device
RAD is based entirely on widely available mobile communications technologies Short Messages
(SMS) and Cell Broadcast Messages (CBM). SMS Based alerting is used to activate selected or
individual RADs, while the CBM is used to activate all RADs. These terminal devices are stand-alone
units that incorporate remotely activated alarms, flashing lights, a broadcast FM radio receiver to be
turned off or on as directed by the message, the displaying of the SMS messages on LCD panel, a self-
test button, message acknowledgement and a dynamic hotline GSM call-back feature for user to
acquire additional information. Five push button switches labeled as Call, Ack, LCD, Test, and Radio
control the operational states of the device. The GSM Alarm Device is a product of the University of
Moratuwa Dialog9 Communication Research Lab.
Figure 33 Circuit level system architecture of the RAD
The Microcontroller and the GSM module are the key components of the Alarm Device. The system
components are illustrated in Figure 33. The microcontroller houses a multitude of peripheral devices
such as internal program flash memory, Data Memory, general purpose I/O, and USARTS. Once in
9
Dialog Telekom url on the Disaster and Emergency Warning Network –
http://www.dialog.lk/en/corporate/cr/ourapproach/innovationinclusion/dewn.html
85
operation, the GSM module listens for any incoming SMS messages or CBMs. CBM-based warning
messages will be broadcast on a predetermined dedicated logical broadcast channel. Upon the
reception of a CBM or an SMS, a notification will be sent by the GSM module to the microcontroller.
The Microcontroller in turn will read and processes the message. If the message is from an authorized
source (in case of SMS) and conforms to a given format the Alarm Device will be triggered. The RAD
is designed to power up from the mains supply but is equipped with a seven hour back up battery as a
secondary power source.
HazInfo project was the first to field test the RADs. Hence usability and effectiveness of the devices
were questionable from the beginning. Having disregarded the issue faced by congestion in SMS
applications the units were found to be exceptionally reliable during Lab tests. The external turning-
indicator type flashing light used in vehicles and 40watt speaker ranked the RAD to be the unit to have
the most forceful active alerting functions. The 160 character limited display designed for English text
messaging only established the device to have very low effectiveness. Since the GSM signals cover
only 60% of Sri Lanka the units would be limited to locations with good signal coverage. The sets had
absolutely no value addition to integrating them in to the village daily life, reducing the effectiveness
scores further.
J2ME Applet embedded Mobile Phone
The research used Nokia 6600 MOP that are powered by a 104MHz ARM processor, and is based
around Symbian’s Series 60 platform. Microimage10 developed a J2ME applet that sits on Symbian
Operating System. The MOPs are activated by a SMS sent from an Internet Application that can be
configured to send alerts to all or a group of MOP handsets. The GSM Java enabled SMS mobile
phones receive text alerts in Sinhala, Tamil and English, sounds an alarm, and has a hotline GSM call-
back feature.
Given that the MOP had java based client software that fulfilled all three features of the active alerting
function, as well as text messaging in all three national languages the devices was expected to perform
the best with very high effectiveness. Since SMS messaging requires less signal strength opposed to
voice the MOP based solution was, further, trusted to be highly effective. The ubiquitous technology
with a device that can be alongside the ICT-G at all times and could be easily integrated in to their
daily village life would prove to be most reliable device of the lot.
Operational States of the GSM devices
Power On -- the RAD is powered by 240v 60Hz AC and when the main power is down it is run by a
Matrix 7v DC Lithium Battery. The Nokia 6600 has a rechargeable 9v Battery. At this point the GSM
Module in both devices listen to SMS of the type that are Alerts and be on Standby. SMS Alert is a
specific SMS with a known header that triggers the device to initiate the Alerting Sequence. The total
message is the size of 2 SMS messages; i.e. 2 x 140 7bit characters. Siren On, Light On, and Text
Display are instantaneous. The User can press the “Acknowledge button” to turn the Siren Off and
10
Microimage website – http://www.microimage.com
86
Light off. The User then uses GSM Telephony by pressing the “Callback button” to dial the Alert
Sender for direct conversation or to receive a voicemail of the CAP message. Final step is pressing the
Radio On to listen to an FM Emergency Broadcast Station (not available in the Nokia 6600). This
completes the cycle of operational states of the 2 GSM based Terminal devices. Figure 34 shows all
the operational states of the MOP and RAD.
Siren On Light On
SMS Alert Text Display Siren Off
Telephony Light Off
Power On Standby Radio On
Power Off Testing Radio Off
Figure 34 Expected operational states of the RAD & MOP
Encoding and Transmission of CAP Messages
CAP message is entered via an Internet based application: Disaster and Emergency Warning Network
(DEWN); namely a HTTP Software Application. The HTTP server strips the <msgType> and
<description> elements of the CAP message. The data is then transformed and packed in to 2 SMS
packets and delivers to the Short Message Service Controller (SMSC). This truncated CAP message is
then forwarded to the GSM Terminating device: RAD or MOP via the Dialog GSM network.
Reception and Decoding of CAP Messages
The first 140 character 8 bit special SMS text message contains a header code of 10 characters, which
is decoded by the RAD’s Microcontroller and by the MOP’s J2ME Applet. Remaining 130 characters
contain the text message taken from the <msgType> and <description> of the original CAP message.
The software in the devices activates the alarms of the devices and display the text tripped from the
specially encoded SMS packets. The RADs displays only the English portion of the text message;
where as the MOPs display the Sinhala, Tamil, and English text messages.
87
25.3.Solana Networks/Innovative Technologies VSAT/IPAS
VSAT System
VSAT Technology is predominantly used in Sri Lanka for high speed 2-way data exchange, especially
satellite Internet access. The LM-HWS uses a star network topology to connect the Sarvodaya
Community Disaster Management Center’s Hazard Information Hub (HIH) in Moratuwa, serving as
the central gateway to interconnect earth stations in Urawatta and Hambantota that serve as nodes.
However, the setup also allows any location to serve as the central gateway and, thereby, serving as a
back hub site.
The Antenna used at the HIH is a CHANNEL MASTER 2.4 M C-Band linear antenna. The other
ground stations use 1.8M C-Band linear antennas. Each antenna are aligned with AsiaSat and are fitted
with an AGILIS C-Band BUC and LNB. Two co-axial cables running from the BUCs and LNBs are
connected to iDirect 3000 series modems/routers and local network switches.
Internet access via the VSAT system is provided by Speedcast Limited (Speedcast). Headquartered in
Hong Kong, Speedcast offers satellite based value added telecommunications services to over 25
countries in the Asia-Pacific region. Speedcast has its regional Network Operations Centre (NOC)
located in Hong Kong. The Speedcast NOC is equipped with performance monitoring and network
management systems to facilitate problem resolution and is staffed with engineers working 24 hours, 7
days a week, 365 days a year to ensure continuity of services.
Figure 35 Rear of BUC ALB and Front/Rear of iDerct Modem
IPAS Alerting Software Application
Although not CAP-compliant in its present form, IPAS possesses some similar features to CAP. Like
CAP-based systems, the IPAS GUI is accessed through a web-based form. The authorized sender logs
in and can then select the “Issue Alert” page to begin composing an alert message. Among the features,
users can set the send time, duration of the alert, geographic target area, type of message, level of alert,
and compose an alert message. The message box feature is equivalent to the message description in the
CAP template.
The IPAS Server GUI also allows authorized users to issue hierarchical alerts to subscribers via the
Internet. That is to say, alerts can be issued on a national, regional or community wide basis. Further,
IPAS alerts can be issued according to a scale of severity from 1 to 5 (5 being the highest severity).
88
For receiving alerts, IPAS utilizes an alert window, visual and audio approach. Users download and
install a Java applet on their computers and then register for the types of emergency alert notifications
that they want and can filter them according to geographic scope and severity. In the event of an
emergency alert broadcast, the alert message will pop-up in a window on the subscribed user's
computer screen with an audible alarm.
IPAS Computer Operational States
IPAS Message Clients were always tested on a Desktop or Notebook computer (PC). The computers
were preloaded with Windows XP or Windows Vista Operating Systems. The software is configured
to be activated upon PC power up. Thereafter, the IPAS Client prompts the user to Log In. After
logging in, the IPAS Client runs as a service in the background and is on Standby listening to dedicated
communication ports for any alerts messages. When an IPAS Alert is received, if configured to Siren
On , immediately sends a siren sound to the PC speakers and a Text Display window pops up on the
screen with an Alert Message. After 10 seconds the sound is set to Siren Off. The end-users can cancel
the Text Display and set the IPAS Client back on standby mode. If the user does not wish to receive
alerts then the user can Log Out. Also if the PC is being powered down the operating system will force
IPAS to Log Out before turning the PC Power off.
IPAS Alert Siren On
Power On Log In Standby Text Display
Power Off Log Out Siren Off
Figure 36 Operational States of the IPAS on a Personal Computer (PC).
25.4.Sri Lanka Telecom Nomadic Telephony System
CDMA Nomadic Phone
The original intention of the project was to use fixed Public Switch Telephone Network (PSTN) lines.
However, the infrastructure and capacity does not exist in the rural areas if Sri Lanka; i.e. fixed line
network has not rolled out in to rural Sri Lanka. Therefore, CDMA FXP was used in the research, a Sri
Lanka Telecom11 solution called City Link for telephony mainly for Rural Sri Lanka. The phone sets
11
Sri Lanka Telecom website – http://www.slt.lk
89
also provide 1xRTT capabilities with Voice, SMS, Fax, and Internet connectivity (54Kbps), which
transmit on the 8.5 KHz and 18.5 KHz frequency spectrums. CDMA Fixed Wireless Phones with
built-in speakerphones to provide voice communication via the PSTN FXPs.
The alerting for was a simple voice call, which would allow for an alert message to be voiced in any
language. However, the language diversity of the staff at the HIH would limit the multilingual
capabilities. With CDMA coverage spanning all most the entire country it would be highly reliable
solution. Loud ring tone is a definite attention getter. However, the nomadic terminal device requires
that the ICT-G be in close proximity. Also the if the call is received by the ICT-G, then
acknowledgment is instantaneous.
The Internet access of the 1xRTT capabilities were tested from a remote location; where a prepaid card
has to be purchased from an SLT outlet and registered through the SLT web portal over the internet.
The dilemma is to access the internet via the CDMA phone one must access the internet via an
alternate mean to registered the prepaid access information. The CDMA Terminal Device is coupled to
a computer via a RS232 serial connector. The TCP/IP must be configured to use “dialup” though a
short code. Once connected the bandwidth varies from 36 – 54kbps depending on the signal strength.
In order to use the 1xRTT capabilities the Terminal Device must be supplied AC power through the
adopter or a 12v DC supply and does not function off the rechargeable battery.
SMS was not tested since the subscribed service package was limited to voice and SMS was not
activated. Therefore, the project did not have the opportunity to test alerting over SMS to the FXPs.
Operational States of CDMA 2000 1xRTT Device
In order to power up the CDMA2000 1xRTT terminal device, the unit must be plugged in to the main
AC power or the battery must be charged and have minimal capacity. The signal strength must be at
least 2-3 bars in order to make a voice call. The standard phone ringing tone gets the attention of the
call recipient. The unit has the option of setting 3 different ring tones. When the phone rings (i.e. a call
connection being requested) the user picks up the handset receiver with microphone and speaker to talk
to the calling party on the other end of the connection; once call is complete either the calling party or
receiving party can terminate the call. The terminal unit is on standby waiting to receive a call. The
unit can also receive SMS when on standby mode. When a text message is received a ring tone alerts
the recipient that a message has arrived. The user can view the message, reply, delete, or navigate away
from the SMS display. The user can connect to the internet. While being connected the FXP unit can
receive both SMS and Voice calls. Similarly, when on a voice call the FXP can receive a SMS as well
as stay connected to the internet. When running on battery power, if the battery capacity is below the
minimum charge level, the unit will automatically power off.
90
Push SMS Get SMS Ring
Power On Standby Cancel Talk
Power Off Internet Dial
Figure 37 Operational states of 1xRTT CDMA 2000 Unit
91
26.TECHNICAL ANNEX 15: GUIDELINES FOR THE HIH

26.1.Overview
The LM-HWS is a system for creating and issuing alert messages through the Disaster Management
Institute Hazard Information Hub (HIH) to villages belonging the Sarvodaya’s Sri Lanka network. It is
important to note that alerts issued by the LM-HWS are not “public” alerts. Instead, the HIH issues
alerts to members of a closed user group. The members of the closed user group are designated “first
responders” that have been given training by Sarvodaya Shanti Sena (Peace Brigade) and have been
trained and certified in conjunction with the Last Mile Hazard Information Dissemination Project.
Local first responders are members of the local community and it is they, or their authorized
designates, who are responsible for determining if a local, community-wide (village) warning is to be
issued. This flow of communications means that the LM-HWS is a two-stage relay network that
disseminates information to alert local first responders. These guidelines attend to the First Stage of
the LM-HWS only. Training and response plans for local villages are intended to deal with the Second
Stage of the system. Figure 38 provides a schematic illustration of the LM-HWS as a two stage relay
network.
Hazard Event
Hazard
Information Hub
First Stage
Local First
Responders
Second Stage
Villagers,
Community
Figure 38 The LM-HWS depicted as a two-stage relay network
92
26.2.The LM-HWS and its Responsibilities for Issuing Alerts
The HIH and the government
In the event the government of Sri Lanka issues a public warning, the Hazard Information Hub (HIH)
will relay this message directly through its network. Local first responders will act on the message as
they see fit, but these actions should be based on local response plans and any instructions provided by
the government in the initial message or in subsequent public communications (e.g., official messages
broadcast by the media).
If the government does not issue CAP-compliant messages, the HIH will need to convert these
messages into the CAP format quickly and accurately. If government messages are issued in a
standard protocol other than CAP it might be possible to automate the conversion process.
However, in no case should staff at the HIH modify or otherwise revise the contents government
message, except to ensure that it is capable of being relayed over the LM-HWS network. In certain
cases, the HIH Executive (see below) may direct an Authorized User at the HIH to issue
supplementary messages in conjunction with official government warnings.
It is recommended that the optional CAP element <description> be used to relay government warnings
in their entirety—if possible, copied and pasted into a CAP message—rather than asking HIH
personnel to interpret or to transcribe the message and risk introducing errors in the transcription
process.
The HIH and local first responders
The HIH will not issue messages that provide specific instructions to local first responders, except
those that might be relayed directly from the government.
Instead, all HIH-originated messages should contain enough information to enable a local first
responder to decide if s/he should activate their local response plan based on their training and
judgment.
It is the responsibility of the HIH to ensure that all messages sent through the LM-HWS are authorized
and conform to an agreed upon CAP Profile Document. These guidelines also establish a CAP Profile
Document for the LM-HWS.
Any proposed changes made to the CAP Profile Document that might affect the warning message, or
its display on any communications device, must be discussed with training personnel and others
involved in the First Stage relay network (including service providers and technology partners).
In the longer term it is anticipated that local first responders will maintain a state of observational
readiness and provide information updates to the HIH. These updates will include local reports of
hazard impacts or other environmental indicators will serve to provide an upstream flow to support
situational awareness at the HIH.
93
It is also anticipated that a network of automated sensors placed in local communities and other
strategic locations will provide some of these upstream flows. In all cases, it is expected that both
downstream and upstream messages will be CAP compliant.
Further details on the upstream component will be developed in due course and attached as an
addendum to these guidelines.
Local first responders and their communities
It is the responsibility of local first responders to maintain a state of readiness in order to be able to
receive messages from the HIH. This state of readiness includes a number of responsibilities. Local
first responders must
 Ensure that their local communications equipment is in good operating condition and that
battery-operated equipment is adequately charged.
 Ensure that their local communications equipment is “on” and ready to receive messages from
the HIH on a 24x7 basis.
 Ensure that at least one authorized person is responsible for monitoring the local
communications channel(s) for incoming messages at all times.
It is also the responsibility of local first responders to decide on appropriate action when a message is
received. Appropriate action will depend on the information content of the message and the local
community response plan. For example, an urgent tsunami warning for the west coast of Sri Lanka
might prompt some local villages to activate their evacuation plans while in neighboring villages they
activate plans in preparation to receive evacuees.
Again, the HIH will not issue specific instructions to each community but will instead strive to include
enough information in its messages to enable local first responders to instigate appropriate action with
respect to local response plans.
94
26.3.Staffing and Training at the HIH
The following chart provides an overview of the organizational structure of the HIH.
E x e c u t iv e
C o o r d in a t o r
A s s is t a n t
M o n it o r s S u p p o rt
Figure 39 Organizational structure of the HIH
The HIH Executive is responsible for taking decisions to issue or not issue warning messages from the
HIH. Members of the executive are senior managers or directors from key stakeholder groups.
The HIH Coordinator is responsible for day-to-day operations of the HIH, as well as for taking a lead
role when a decision has been taken to issue warning messages. The Coordinator will be qualified as
an “Authorized User” and has the authority to issue warning messages when first approved by the
Executive.
The Assistant supports the Coordinator in his/her responsibilities and may or may not be qualified as
an Authorized User of the system.
The Monitors are those staff members responsible for tracking and recording details from the various
information feeds provided at the HIH. The HIH will be staffed on a continuous basis by at least one
Monitor. Monitors will normally also be qualified “Authorized Users,” meaning that they have the
authority to compose and issue warning messages when a request to issue a warning is first approved
by the Executive.
Support staff include those responsible for upkeep of technology and other equipment at the HIH, as
well as those who might be called in to the HIH to provide additional assistance during emergency
events (e.g., telephone support) or on an as-needed basis (e.g., during exercises). Support staff may or
may not be qualified as “Authorized Users.”
95
Initiating a warning message
The following flowchart describes the simplified procedure for initiating warning messages from the
HIH.
EOI Mon itor verifies

incoming data
C lose ca s e; rem ain

vigilan t; obt ai n mo re
Mon itor consu lts with
info rmation .
ex ecutive
Decision to send
messa ge
No
Yes
Mon itor compo se s

messa ge using pre-
defined p rocess
Mon itor initiates

messa ge and no tifies
HIH coo rdin ator
Figure 40 Simplified procedure for issuing warning messages from the HIH
96
Step 1: Event of interest (EOI)
When a staff member receives information that is potential cause for concern, it is first verified by the
staff member and the source, time, and other pertinent details are recorded in the HIH database as an
“event of interest” (EOI). See Annex B for guidelines to documenting an EOI.
The HIH database automatically assigns an incident identifier to the entry even if it is eventually
decided not to issue an alert message. In this way, all events of interest are assigned unique identifiers
for recordkeeping purposes.
In the event an alert is actually issued by the LM-HWS, the incident identifier can be used to collate
multiple messages that refer to different aspects of the same incident. For example, if a large
earthquake is reported off the coast of Indonesia and the HIH staff member decides it is an event of
interest (EOI), then the details are entered into the HIH database and the event is assigned a unique
identifier compatible with the CAP <incidents> element.
No other CAP elements are specified until a decision to issue an alert has been made.
If an alert is issued based on the EOI, then the assigned unique identifier serves as a common referent
for all subsequent messages based on that event, including messages of varying priorities targeted for
different communities, as well as initial alerts and any follow up messages.
The EOI is terminated on a decision by the HIH Executive to “close” the case. When this occurs the
unique identifier assigned to the EOI is retired. Subsequent EOI are assigned new unique identifiers
compatible with the CAP <incidents> element.
Step 2: Consultation with HIH executive
Having recorded the event of interest (EOI) in the HIH database, the staff member then contacts a
member of the HIH executive by telephone.
It is imperative that a current and complete list of telephone numbers of the members of the Executive
is maintained and available the HIH for this purpose. The current list of contacts is provided as Annex
A to these guidelines. Annex A should undergo regular review and updates to ensure the contact list
remains current.
The staff member will provide details of the EOI to the members of the Executive and conduct any
follow up inquiry as directed by the Executive.
The members of the Executive will take a decision as to whether to initiate a warning message and any
related details pertaining to that message, such as geographic specificity, priority level, special
instructions, and so forth.
97
Step 3: Decision to send message
Having reached a decision, the members of the Executive will provide details to the staff member who
will then enter those into the HIH database under the incident identifier established for that EOI:
1. Time that a decision was requested of the Executive.

2. Executive member(s) with whom the EOI was discussed
3. Time that the Executive issued a decision.
4. Details of the decision and/or further instructions from the Executive.
Step 4: Message composition
Having entered the details of the Executive’s decision into the HIH database, the staff member will
then take immediate action based on the details of the decision.
The following points indicate possible scenarios:
• Decision not to issue a warning and close the case (EOI is terminated).
• Decision not to issue a warning but to remain vigilant (EOI remains open).
• Decision not to issue a warning but to obtain more information and report back to the Executive
(EOI remains open).
• Decision to issue a low priority warning and to remain vigilant.

• Decision to issue a medium priority warning, to obtain more information and report back to the
Executive.
• Decision to issue a high priority warning and to provide continuous updates to the Executive
and/or the HIH Coordinator.
If the staff member has been directed to issue a warning then it is important that they follow the pre-
defined process for composing messages in conjunction with any steps defined by the CAP interface
application. If the staff member does not have the authority to issue a warning message, then they
must contact an “Authorized User” immediately to complete this step.
In cases of extreme urgency, the HIH Coordinator may grant to another staff member temporary
authority to issue a warning on their behalf. For example, if there are no Authorized Users on-site at
the HIH during a critical moment, a staff member may obtain verbal authorization from the
Coordinator to issue a warning (provided that the Executive has been consulted and has approved the
warning).
See the following sections on “Training” “Activation” for further details regarding Authorized Users
and message composition.
The resulting CAP message will be assigned (either automatically or manually) the same unique
incident identifier as the EOI file that spawned the decision to issue a warning.
98
In the case that there is a discrepancy or any uncertainty on the part with regard to message
composition (e.g., details of content) then the Authorized User must contact the HIH Coordinator or
HIH Executive immediately for advice before issuing the message.
Step 5: Issuing a Message
Having composed the warning message using the CAP interface application, the Authorized User will
issue that message immediately by following the appropriate procedures to ensure it is presented to all
designated channels (i.e., telephone, VSAT, ADSR, etc.).
The Authorized User will then immediately contact the HIH Coordinator by telephone to report that a
message(s) has been issued.
Any known problems or delays must be reported to the HIH Coordinator immediately.
When a message is issued over the LM-HWS the Authorized User is required to complete a post-
activation report. See the following section “Post-Activation” for guidelines concerning this
requirement.
Training
It is recommended that an authorization procedure be developed for HIH staff members to qualify
them as “Authorized Users” with the authority (and related responsibilities) to compose and issue
warning messages over the LM-HWS.
All new LM-HWS personnel should be required to attend a training session provided by the
HIH/Sarvodaya. In addition all authorized users should be required to attend regular refresher training
to maintain their status. Regular training sessions could be scheduled prior to the cyclone season in
November. In addition, authorized users should be required to complete monthly practice sessions
composing and initiating messages using the system. To promote compliance, the HIH computers
could be designed to log these monthly sessions.
Each initial and refresher training session should include:
 An LM-HWS presentation that explains the importance of the system and how it is intended to
operate.
 A detailed description of activation criteria and procedures for the LM-HWS.
 An overview of likely risks and hazards in the communities, histories of disasters, and examples of
activation scenarios.
 Training in working with information technology at the HIH.
 Training in interpretation of the data feeds at the HIH.
 Demonstration of LM-HWS activation.
 Hands-on practice in recordkeeping and message composition using both a practice sheet and with
99
the actual software.

 Testing and certification.
Upon completion of the training session, candidates should be asked to sign an “HIH Authorized User”
form, which serves as a contract to bind the staff member to the HIH Guidelines. This contract should
address a number of matters related to the operational and security dimensions of the HIH:
 Authorized users should agree to be responsible for the security of any passwords or access codes
or keys issued to them.
 Authorized users should agree to complete monthly practice sessions on using the system (these
could be logged on the HIH computer system).
 Authorized users should agree to attend regular refresher training courses and other professional
development as directed by the HIH Coordinator.
 Authorized users should agree to follow established HIH procedures and guidelines for monitoring
data feeds and “on call” responsibilities.
 Authorized users should agree to follow established HIH procedures and guidelines for composing
warning messages.
 Authorized users should agree to follow established HIH procedures for initiating warning
messages.
 Authorized users should agree to disclose all actions taken, communications given and all other
forms of information pertinent to the use of the system should they activate it for any reason.
26.4.Monitoring
The HIH maintains data feeds from a number sources. Duty personnel are responsible for ensuring
that these sources are active and to immediately report any problems to the HIH Coordinator.
The HIH will be staffed on a 24/7 basis, with the Coordinator and at least one member of the Executive
remaining “on call” at all times.
An active set of data feeds will be maintained at the HIH to identify potential events of interest across
a range of hazards. The LM-HWS is intended as an “all-hazards” system and the following list
provides a range of events that should be included among the data feeds:
• Hazard
• Tsunami
• Cyclone
• Volcano (ash fall)
• Flooding
• Landslide
• Dam break
• Coastal/marine events
• Fire
• Industrial/radiological
• Civil disturbances
100
In addition to its data feeds, the HIH is to maintain a current list of contact information of secondary
sources that might be helpful to confirm or otherwise obtain further details on any Events of Interest.
The HIH Coordinator is responsible for the upkeep of this list and it should be reviewed regularly.
The current list of secondary source contacts is found in Annex C.
The HIH should also maintain one or more incoming telephone lines (“hotlines”) to receive reports
from outside agencies or organizations (including local communities that might have observed a
hazard event). This number should not be public but instead be made available to those agencies and
community groups that are likely to be able to provide advance notice of a hazard event (e.g., Pacific
Tsunami Warning Centre, local first responders).
The current hotline telephone number is listed in Annex D.
Duty Procedures
The HIH should develop a checklist and procedures for duty personnel who are responsible for
monitoring the data feeds. This list should include the following items:
• Ensure all communications equipment (incoming/outgoing) is in good working order.

• Establish a procedure for conducting a regular “sweep” of all primary data sources” (perhaps
2-3 times per hour).
• Identify any recordkeeping requirements (equipment out of order, report on data feeds, etc.)
• Establish a procedure for temporary replacement/substitution of duty personnel (e.g., sick
days).
• Establish a procedure for off-site monitoring of data feeds (if applicable).
• Establish a procedure for hand-off of duties during shift change.
Off-duty personnel
Off-duty personnel may be called in to provide support at the HIH in the event a warning message has
been issued and a hazard impact is expected. This decision will be at the discretion of the HIH
Coordinator.
26.5.Activation
Message Priority
When reporting an event of interest (EOI) an authorized user may request the HIH Executive to issue
an Urgent Priority warning message when one or more of the following threat conditions are present:
101
 The life or safety of groups, communities or villages is at immediate risk.

 The danger to the community is impending and widespread.
 The potential impact to the community is catastrophic.
Local first responders needs to be informed of critical, life saving information and be advised to
activate their local response plans.
The following table contains recommended CAP values for <urgency>, <severity>, and <certainty>
elements in an urgent priority message.
Table 16 CAP values for an urgent priority message

CAP <info> element Value (recommended) Interpretation
Urgency “Immediate” or “Expected” Immediate responsive action
should be taken
Severity “Extreme” or “severe” Hazard presents an
extraordinary threat to life or
property
Certainty “Observed” or “likely” The hazard event has occurred
or is ongoing (or, > 50%).
Alternately, an authorized user may request the HIH Executive to issue a High Priority warning
message when one or more of the following threat conditions are present:
• The life or safety of communities or villages is possibly at risk.

• Neighboring communities or villages have been issued an urgent priority warning.
• Residents of the community may see/hear/smell (detect) signs of the hazard and may perceive a
danger or health risk.
• Local first responders need to be informed of the hazard situation to provide information to
community members.
• Local first responders must be advised to standby to activate their local response plans.
elements in a high priority message.
Table 17 CAP values for a high priority message

Urgency “Expected” or “Future” or Responsive action might need
“Unknown” to be taken in near future.
Severity “Severe” or “Moderate” or Hazard presents a significant
“Unknown” threat to life or property.
Certainty “Observed” or “Likely” The hazard event has occurred
or is ongoing (or, > 50%).
102
An authorized user may request the HIH Executive to issue a Low Priority warning message when
one or more of the following conditions are present:
• The life or safety of a community might be at risk due to a developing hazard.

• A neighboring community has been issued a high priority warning.
• Residents of the community may see/hear/smell (detect) signs of a hazard or nearby response
effort and may be curious.
• Local first responders need to be informed of the hazard situation to provide information to
community members.
• Local first responders must be advised to standby for further information.
elements in a low priority message.
Table 18 CAP values for a low priority message

Urgency “Future” or “Unknown” Responsive action might need
to be taken in near future.
Severity “Moderate” or “Minor” Hazard presents a minimal
threat to life or property.
Certainty “Possible” or “Unknown” The hazard event is possible
but not likely (p < 50%).
Message composition
To minimize error and to maintain consistency, authorized users are to compose warning messages
using the standard format and procedure established at the HIH.
All messages issued over the LM-HWS will be compliant with Common Alerting Protocol (CAP
version 1.1), as specified in the OASIS Open Source Standards document.12
CAP standard specifies that alert messages are composed of an <alert> segment with may contain one
or more <info> segments, each of which may contain one or more <area> segments.
Warnings issued by the LM-HWS will be multilingual (Tamil, Sinhalese, and English) and may
contain multiple area segments.
The following diagram shows a recommended structure for CAP-compliant multilingual messages to
be issued by the LM-HWS. Each <alert> will include three <info> segments providing identical
information in Tamil (ta), Sinhalese (si) and English (en). Each of the three info segments will contain
12
http://www.oasis-open.org/committees/download.php/14759/emergency-CAPv1.1.pdf
103
identical <resource> and <area> segments, translated if necessary.
alert
<incidents>
<identifier>
<sender>
<sent>
<status>
<msgtype>
<scope>
info info info

<language> ‘ta’ <language> ‘si’ <language> ‘en’
<category> <category> <category>
<event> <event> <event>
<urgency> <urgency> <urgency>
<severity> <severity> <severity>
<certainty> <certainty> <certainty>
<description> <description> <description> resource
<resourceDesc>
area
<areaDesc>
Figure 41 CAP message structure for LM-HWS
To ensure CAP compliance, all messages should be composed using the CAP interface tool, with the
required elements specified in the following tables.
Table 19 CAP <alert> sub-elements

<Alert> sub-element Description Implementation remarks
<incidents> Unique identifier of the EOI This is assigned prior to
decision to issue warning and
is a common referent for all
messages issued with respect
to the same EOI.
104
<identifier> A unique identifier of the Each message has a unique

message. identifier, but different
messages can refer to the same
<incidents> reference.
Automatically assigned by the
CAP software
<sender> Identifies the originator of this Should specify HIH as

message. organization and the
Authorized User that issued
the message. Assigned
automatically based on the log-
in ID of the Authorized User.
<sent> Date and time the message was Automatically assigned by the
transmitted. CAP software at the moment
the message is transmitted.
<status> Code that denotes appropriate See the section on “Message

handling of the message. Status” for further details.
<msgType> Code that denotes the nature of See the section on “Message
the message Type” for further details.
<scope> Code that denotes the intended See the section on “Message
distribution of the message. Scope” for further details.
105
Table 20 CAP <info> sub-elements

<Info> sub-element Description Implementation remarks
<language> Code that denotes the Two digit ISO 639-2 codes are
language of the info sub- used to specify language of the
element of the message message: ta/si/en
<category> Code that denotes the category See the section on “Message
of the event referred to by the Category” for further details.
message.
<event> Text that denotes the type of See the section on “Message
event. Event” for further details.
<urgency> Code that denotes the time to See the section “Message
impact of the event. Priority” for further details.
<severity> Codes that denotes the scale of See the section “Message
impact of the event. Priority” for further details.
<certainty> Code that denotes the See the section “Message

probability of the event. Priority” for further details.
<description> Text to describe the subject of To be used primarily for
the event. relaying complete text of
government warnings within a
CAP message. Event element
should be “Government
warning.”
Table 21 CAP <resource> sub-element

<Resource> Description Implementation remarks
sub-element
<resourceDesc> Text that describes the type See the section “Message
and content of a resource Attachments” for further
attached to the message such as details.
an audio or image file.
Table 22 CAP <area> sub-element

<Area> sub-element Description Implementation remarks
<areaDesc> Text that describes the affected See the section “Area
area of the alert message. Description” for further details.
106
Message Status
CAP has a required sub-element <status> to specify the appropriate handling of an alert message. The
CAP standard specifies several values for this sub-element, and the following table describes the
recommended implementation for the LM-HWS.
Table 23 CAP message <status> values

Value Implementation remarks
“Actual” To be used when the message refers to a real event and when all
recipients are to take action.
“Exercise” To be used when the message is transmitted as part of a scheduled or

unscheduled exercise. An exercise identifier could be added to the
optional <note> sub-element in the CAP message.
“System” Not to be used for LM-HWS.
“Test” To be used when testing internal technical parameters of the system.

Recipients are to disregard the message. Messages that specify “test” in
the <status> element should not be relayed to local first responders.
Software filters should be designed to recognize these messages as such
and to prevent them from being relayed to local first responders.
“Draft” Refers to templates that can be stored on a HIH database to expedite

message composition. Software filters at the HIH should be designed to
prevent the transmission of messages that specify “Draft” in the <status>
element.
107
Message Type
CAP has a required sub-element <msgType> to specify the nature of an alert message. The CAP
standard specifies several values for this sub-element, and the following table describes the
recommended implementation for the LM-HWS.
Table 24 CAP message <type> values

“Alert” To be used for initial information only.
“Update” To be used to indicate new information. The message content of an
update supersedes that of earlier messages.
“Cancel” To be used to cancel earlier messages.
“Ack” Not to be used for LM-HWS.
“Error” Not to be used for LM-HWS.
Updates and cancellations can specify earlier messages identified in <references> if this optional sub-
element is implemented in the system.
The sub-element <references> provides a place for an extended message identifier (in the form sender,
identifier, sent) to specify earlier messages that might be issued with respect to an EOI.
For example, if a tsunami event were to trigger a low priority message for one region and a high
priority message for another region, the <references> sub-element can refer to the extended identifier
to ensure the specificity of updates and cancellations.
Message Scope
CAP has a required sub-element <scope> to denote the distribution of an alert message. The CAP
standard specifies several values for this sub-element but because the LM-HWS is a closed network
only the “Restricted” value should be implemented.
Table 25 CAP message <scope> values

“Public” Not to be used.
“Restricted” To be used to specify a closed user group of recipients. Designated local
first responders represent this closed user group.
“Private” Not to be used.
In future, if the LM-HWS becomes integrated with larger systems, the optional <restriction> sub-
element can provide addressability to distinguish between different user groups within the network.
108
Message Category
CAP has a required sub-element <category> to denote the category of the subject event of an alert
message. The CAP standard specifies several values for this sub-element, and the following table
describes the recommended implementation for the LM-HWS.
Table 26 CAP message <category> values

“Geo” To be used for geophysical hazard events, such as earthquakes, tsunamis,
landslides.
“Met” To be used for weather hazard events, such as cyclones, flash floods,
lightning.
“Safety” To be used for general emergency and public safety.
“Security” To be used for civil disturbances.
“Rescue” Not to be used.
“Fire” To be used for fire hazard events.
“Health” To be used for medical or health hazard events (e.g., disease outbreak,
water contamination).
“Env” To be used for pollution or other environmental hazard events (e.g., air
quality).
“Transport” Not to be used.
“Infra” To be used for infrastructure hazard events (e.g., dam failure).
“CBRNE” Not to be used.
“Other” To be used for other events.
The CAP standard specifies that multiple instances may occur within a single <info> block, which
means that several values may be used to classify an event.
It is not clear at this point in time as to how important this sub-element will be for the LM-HWS or for
information management at the HIH. However, it is strongly recommended that a value for the sub-
element be included in all messages sent from the HIH because it is required in the CAP standard.
Message Event
CAP has a required sub-element <event> to denote the subject event of an alert message. The CAP
standard does not specify values for this sub-element but the following list provides basic terminology
for naming events and suggested <category> for each, including a special event for government
warnings.
Table 27 CAP message <event> values

Event Category
Boil water warning Health
Child abduction Other
Civil danger Security
109
Coastal flood Met

Contagious disease outbreak Health
Dam break Infra
Cyclone Met
Earthquake Geo
Flash flood Met
Flood Met
Food contamination warning Health
Government warning Other
Hazardous materials warning Env
High wind Met
Landslide Geo
Missing person Other
Power outage Infra
Problem animal Other
Special weather statement Met
Severe thunderstorm Met
Tornado Met
Tropical storm Met
Tsunami Geo
Volcano Geo
Wildfire Fire
This list is not intended to be exhaustive. For the sake of consistency and clarity, the HIH should
ensure a reasonable degree of standardization for naming of events.
The event “Government warning” should be used when relaying messages from the government of Sri
Lanka, with the full text of the warning included in the <description> element of the CAP message.
Message Attachments (Resource)
CAP has a required sub-element <resourceDesc> to refer to additional file(s) with supplemental
information related to the <info> element, such as an audio or image file. Multiple occurrences are
permitted in CAP (version 1.1).
Messages composed at the HIH will include audio files recorded in Tamil, Sinhalese and English.
Audio files will be used principally for the satellite radio component of the network but could be used
for telephone or other devices if necessary.
Either of two methods for creating the audio files can be employed:
 A staff member at the HIH composes three audio versions of the message (Tamil, Sinhalese, and
English) and then reads them into a voice-recording device and saving them as one or more digital
files.
110
 An application automatically converts the CAP message into three audio versions (English,
Sinhalese, Tamil) using text-to-voice software, saving them as one or more digital files.
Audio messages will include certain elements set out in the CAP message, effectively matching the
text version of any warning message issued by the HIH. Certain CAP sub-elements may not be
relevant for audio versions.
How the CAP elements will be scripted may vary depending on which method is used to record the
audio files. The following is one example of a script for an urgent priority tsunami warning that is
integrated with some of the CAP required elements.
It is important to note that some of these elements may not translate easily using text-to-voice software
and may need some additional treatment for the audio version (e.g., date and time in the <sent>
element).
Message header:
Your attention please! This is an activation of the Sarvodaya Hazard Warning System. The
following emergency <msgType>alert</msgType> may affect your village. This is an
<status>actual</status>event. Listen carefully and prepare to take appropriate action based on your
village’s response plan.
Message content:
At <sent>0800hrs local time on July 1, 2006</sent> the <sender>Sarvodaya Disaster
Management Centre </sender> has been informed of a large magnitude earthquake off the coast of
Indonesia. This event is <certainty>likely</certainty> to produce a <event>tsunami</event> and
presents an <urgency>immediate</urgency> and <severity>extreme</severity> risk for the
<areaDesc>village of Gurathwa and the Galle District</areaDesc>.
To repeat: an <urgency>immediate</urgency> <event>tsunami</event>

<msgType>alert</msgType> has been issued for the <areaDesc>village of Gurathwa and the
Galle District</areaDesc>.
Message trailer:
This message was issued by the Sarvodaya Hazard Warning System. If your area is affected, it is
important that you take appropriate action based on your local response plan. Listen to local radio
or television for further updates.
Figure 42 Sample script based on CAP elements
Further discussion will need to take place before message scripting, formatting of the audio files, and
transport options for the <resource> element can be finalized.
111
A key concern is whether the resource file is to be included in the CAP message itself <derefUri> or
whether it will reside on a server for subsequent retrieval <uri>. See CAP standard, section 3.2.3 <uri>
and <derefUri> sub-elements.
Area Description
To the fullest extent possible, messages issued by the HIH should clearly specify the area of impact
and CAP has a required sub-element <areaDesc> for this purpose.
For the purpose of the first phase of the LM-HWS this sub-element will be defined according to the
selected districts and villages participating in the pilot study.
Geographical specificity is particularly important when multiple areas may have differing levels of
priority. Therefore multiple messages should be issued by the HIH when it is determined that the event
could have a range of severity values. For example, villages in the direct path of a hazard event may
receive urgent priority messages, while neighboring communities are issued high or low priority
messages.
The table on the following page specifies Districts and Villages to be used in specifying the
<areaDesc> sub-element for the first phase of the LM-HWS.
Messages may be directed to an entire district (e.g., “Ampara District”) or to a specific village within a
district (e.g., “Ampara, Panama North”).
Further geocoding of CAP messages is possible with the <polygon> <circle> and <geocode> sub-
elements. The HIH will need to determine if this method is to be implemented in the LM-HWS.
Table 28 message <areaDesc> values

District Village
Ampara [District]
Ampara Abeyasinghe Pura
Ampara Panama North
Batticalo [District]
Batticalo Nidavur
Batticalo Palmunnai
Batticalo Periyakallar
Batticalo Saturkondagnya
Colombo [District]
Colombo Moratuwella
Colombo Modara
Galle [District]
Galle Brahamanawattha
Galle Indivinna
Galle Urawatha
112
Galle Valhengoda
Galle Venamulla
Hambantota [District]
Hambantota Krinda/Modarapallassa
Hambantota Modaragama
Hambantota Samodhagama
Hambantota Samudragama
Kalmunai [District]
Kalmunai Kalmunai II
Kalmunai Oluville
Kalmunai Ulla 1
Kalutara [District]
Kalutara Diyalagoda
Kalutara Maggona
Kalutara Thalpitiya
Jaffna [District]
Jaffna Munnai
Jaffna Thondamanar
Matara [District]
Matara Kottegoda
Matara Meddhawatha
Matara Mirissa South
Matara Thallala South
Matara Wathegama North
Trincomalee [District]
Trincomalee Thirukadalar
Message Transmission
Having composed the message, the HIH staff member then issues a warning by transmitting that
message over the LM-HWS. Software located at the HIH converts the CAP elements into suitable
transport formats and relays them to the appropriate gateways.
In the event of a known failure in transmission, the HIH staff member is to attempt to reach the
affected areas by alternate means. A list of alternate telephone numbers should be maintained at the
HIH for this purpose.
In some cases, the HIH staff member may need to contact the first responder in a neighboring village
and request that the message be relayed between villages by hand or other means.
113
Repeat Activations
During an incident the danger level to local villages may change over time. If the danger level begins
to increase in any of the following circumstances, additional messages should be issued only after the
HIH executive has provided authorization to do so:
• A potential secondary impact to the first event creates new hazards to the affected area or to
neighboring villages.
• If new information comes to light that suggests the situation is more certain, severe, or urgent
than initially indicated in a previous message.
• If the affected area must be expanded.
All-clear Messages
It is not recommended that the LM-HWS send “all-clear” messages over the system to the local first
responders. Instead, local first responders in consultation with Sarvodaya should develop a local
procedure for event termination.
Post-Activation
Having successfully issued the message, the authorized user will then immediately contact the HIH
Coordinator to confirm the activation and to request further instructions.
The HIH Coordinator will in turn contact the HIH Executive and provide them with a situation report.
The HIH Coordinator will also determine if support staff will be needed to handle follow-up activities.
The Authorized User is required to complete an Activation Report following activation of the LM-
HWS. The report is signed by the HIH Coordinator and submitted to the HIH Executive. Details of
the report should include the following:
• Name of authorized user, position within the HIH

• Name of person in the HIH Executive that authorized the activation.
• Time and date of the activation.
• Details of the activation, including all elements specified in the CAP messages issued (copies
of the CAP messages should be attached to the report).
• Details of subsequent activations.
114
26.6.Prohibited Practices
There are situations when it is not appropriate to activate the LM-HWS. The HIH Executive will
ultimately decide on when to issue a warning message but a number of principles will generally be
followed.
The LM-HWS should not be used for:
• Transmission of messages to the general public.

• The transmission of “all clear” messages.
• General information on disaster related services and response plans. This is to be provided in
response plans held and maintained by local first responders.
• Any circumstance when life or safety is not threatened.
A process for guarding against system misuse or abuse should be put in place at the HIH. All cases of
accidental activation or misuse must be reported immediately to the HIH Coordinator and to the HIH
Executive.
Effective training is the most important safeguard against misuse and abuse in the first instance.
However, when dealing with a case of misuse by an authorized user, the HIH Coordinator should
consider several options depending on the severity of the case:
• Re-training without probation.

• Re-training with probation.
• Suspension of authorized user status.
• Revocation of authorized user status.
26.7.Testing
The LM-HWS must be tested on a regular basis. The following table presents various testing modes
and the CAP <status> values to be used in message composition.
Table 29 CAP status for test categories

Live Test Silent Test
Closed simulation Village-wide Internal HIH test System
with local first simulation with
(no messages interoperability test
responder. community transmitted to local (no messages
members involved. first responders). transmitted to local
first responders).
CAP <status> CAP <status> CAP <status> CAP <status>
“Exercise” “Exercise” “Test” “Test”
Live Tests may be scheduled or unscheduled. It is recommended that all village-wide live tests be
scheduled.
115
Testing Schedule
The HIH will need to determine a schedule of Live and Silent tests for the system.
Reporting Procedure
The HIH Coordinator will submit a “head-end” report immediately following any test of the system.
This report should include details of the test message, reliability of various elements of the system,
problems encountered during the test, other observations.
Local first responders, supported by Sarvodaya, should submit a “reception” report immediately
following any Live Test of the system. This report should include details of the test message (time
received, message details, etc.), reliability of the reception device, problems encountered during the
test, and other observations.
Support partners (e.g., Dialog, WorldSpace) should be encouraged to submit an “interoperability”

report following any Silent Test that involves their network. This report should identify any problems
encountered during the test or other relevant observations.
It is recommended that the HIH Coordinator compile results of the head-end, reception, and system
interoperability reports into a summary report and recommendations for submission to the HIH
Executive within one week of any Live or Silent Test.
26.8.Annex A: Contact Information
HIH Executive
HIH Coordinator
Assistant to the Coordinator
HIH Monitors
HIH Support staff
Technical partners and service providers
116
26.9.Annex B: Guidelines for documenting an EOI
 Incident Identifier: must be compatible with CAP standard for <incident> element
 Staff member ID:
 Date observed: dd/mm/yy
 Time observed: hh:mm (24hr, local time)
 Event description: (see “Event” list if necessary)
 Location of event: (country, region, ocean, etc.)
 Information source: website, telephone call, etc.; include all details of the source
 Estimated time of arrival (urgency): Hours? Minutes?
 Estimated severity: Extreme? Severe? Moderate? Minor? Unknown?
 Has the Sri Lanka government issued a warning for this event? Yes/No/Unknown
 If so, do you have a copy of the complete text of this warning?
 Have any other governments issued a warning for this event? Yes/No/Unknown
 If so, which governments?
 Have you verified the information with a secondary source? Yes/No
 If yes, then indicate the source:
 Time that the Executive was contacted: hh:mm (24hr, local)
 What are the instructions of the Executive?
o Close the case
o Remain vigilant
o Obtain more information
o Issue an alert
 Details of the Executive’s instructions:
 Time that a decision was taken: hh:mm (24hr, local time)
26.10.Annex C: Secondary sources for confirming Events of Interest
This list must be kept up to date and should be reviewed regularly by the HIH Coordinator.
Table 30 Contact database for confirming events

Last Update:
Event type Primary contact (name, Alternate contact (name,
organization, telephone) organization, telephone)
Tsunami
Cyclone
Volcano (ash fall)
117
Flooding
Landslide
Dam break
Coastal marine events
Fire
Industrial/radiological
Civil disturbances
26.11.Annex D: Telephone “hotline” number for contacting the HIH
The following number is to be made available to local first responders and other organizations at the
discretion of the HIH Coordinator.
The intent of the “hotline” is to provide a telephone link to the HIH for providing advance notice of a
hazard event or for other emergency purposes.
HIH hotline: +94 (x)xxx xxx xxx
118
26.12.Annex F: CAP alert message example
The following is a speculative example of an urgent priority tsunami warning issued using the LM-
HWS CAP profile in the form of an XML message.
<?xml version = “1.0” encoding = “UTF-8”?>

<alert xmlns = “urn:oasis:names:tc:emergency:cap:1.1”>
<incidents>120321072006</incidents>
<identifier>HIH0001</identifier>
<sender>nwaidyanatha@hih.sarvodaya.org</sender>
<sent>2006-07-21T13:40:01+05:30</sent>
<status>Actual</status>
<msgType>Alert</msgType>
<scope>Restricted</scope>
<info>
<language>en</language>
<category>Geo</category>
<event>Tsunami</event>
<urgency>Immediate</urgency>
<severity>Extreme</severity>
<certainty>Observed</certainty>
<description>A large magnitude earthquake off the east coast of Indonesia occurred at 12:00hrs
local time generating a tsunami wave that is expected to reach the coast of Sri Lanka by 13:30hrs local
time. The tsunami wave presents an immediate and extreme hazard to all coastal regions of Sri
Lanka</description>
<resource>
<resourceDesc>audio file (mpg)</resourceDesc>
<uri>http://www.hih.sarvodaya.org/getalertaudio_en</uri>
</resource>
<area>
<areaDesc>All coastal regions of Sri Lanka</areaDesc>
</area>
</info>
<alert>
119
27.TECHNICAL ANNEX 16: TRAINING – HAZARDS
Goal (Session 1)
To impart an understanding of the concepts of hazards, multiple hazards and disaster and to develop
the capability for hazard assessment
Learning Objective
 After completing this session, the participant will be able to perform a hazard identification and
assessment for a selected community
Learning Outcomes
Participant will learn to --

 Distinguish between the concepts of hazard, hazard event, secondary hazards, multiple hazards
and disaster
 Classify and describe types of hazards
 Explain hazard characteristics such as magnitude, frequency, intensity and rate of onset and their
importance
 Conduct hazard identification, hazard assessment and hazard mapping and explain their functional
value
27.1.Definition of Hazards
Standards Australia (2000) defines a hazard as - ‘A source of potential harm or a situation with a
potential to cause loss.’
Another definition can be - A natural event that has the potential to cause harm or loss. A Hazard is a
threat. A future source of danger. It has the potential to cause harm to --
People - death, injury, disease and stress
Human activity – economic, educational etc.
Property - property damage, economic loss of
Environment - loss fauna and flora, pollution, loss of amenities.
Some examples of hazards in Sri Lanka are cyclones, floods, landslides, droughts, tsunami and
lightning.
120
27.2.Hazard Events
Environmental events become hazards once they threaten to affect society and/or the environment ad-
versely. A physical event, such as a volcanic eruption, that does not affect human beings is a natural
phenomenon but not a natural hazard. A natural phenomenon that occurs in a populated area is a haz-
ardous event.
A hazardous event that causes unacceptably large numbers of fatalities and/or overwhelming property
damage is a natural disaster. In areas where there are no human interests, natural phenomena do not
constitute hazards nor do they result in disasters.
Magnitude is an important characteristic for analyzing hazards since only occurrences exceeding some
defined level of magnitude are considered hazardous.
 The level of harm is governed by
 Magnitude of the hazard
 Frequency of hazard or recurrence
 Intensity at the impact point
27.3.Multiple Hazards
When more than one hazard event impacts the same area, there arises a multiple hazard situation.
These different hazard events may occur at the same time or may be spaced out in time. The planning
process in development areas does not usually include measures to reduce hazards, and as a conse-
quence, natural disasters cause needless human suffering and economic losses. From the early stages,
planners should assess natural hazards as they prepare investment projects and should promote ways of
avoiding or mitigating damage caused by hazards. Adequate planning can minimize damage.
27.4.The Return Period
Majority of hazards have return periods on a human time-scale. Examples are five-year flood, fifty-
year flood and a hundred year flood. This reflects a statistical measure of how often a hazard event of a
given magnitude and intensity will occur. The frequency is measured in terms of a hazard’s recurrence
interval.
For example, a recurrence interval of 100 years for a flood suggests that in any year, a flood of that
magnitude has a 1% chance of occurring.
Such extreme events have very low frequencies but very high magnitudes in terms of destructive
capacity. This means that an event considered being a hundred year flood would cause severe damage
compared to a five-year flood.
121
27.5.Classification of Hazards
Are hazards natural?
There are many different ways of classifying hazards. One is to consider the extent to which hazards
are natural. Natural hazards such as earthquakes or floods arise from purely natural processes in the
environment. Quasi-natural hazards such as smog or desertification arise through the interaction of
natural processes and human activities.
Technological (or man-made) hazards such as the toxicity of pesticides to fauna, accidental release of
chemicals or radiation from a nuclear plant. These arise directly as a result of human activities.
A typology based on Hewitt and Burton (1971) would appear as follows.

1. Atmospheric Atmospheric
Single element Combined elements/events
Excess rainfall Hurricanes
Freezing rain (glaze) ‘Glaze’ storms
Hail Thunderstorms
Heavy snowfalls Blizzards
High wind speeds Tornadoes
Extreme temperatures Heat/cold stress
2. Hydrologic 3. Geologic
Floods – river and coastal Mass-movement
Wave action Landslides
Drought Mudslides
Rapid glacier advance Avalanches
Earthquake
Volcanic eruption
Rapid sediment movement
4. Biologic 5. Technologic
Epidemic in humans Transport accidents
Epidemic in plants Industrial explosions and fires
Epidemic in animals Accidental release of toxic
Locusts chemicals
Nuclear accidents
Collapse of public buildings
Natural hazards and human intervention
Although humans can do little or nothing to change the incidence or intensity of most natural phenom-
ena, they have an important role to play in ensuring that natural events are not converted into disasters
by their own actions. It is important to understand that Human intervention can increase the frequency
and severity of natural hazards.
122
For example, when the toe of a landslide is removed to make room for a settlement, the earth can move
again and bury the settlement.
Human intervention may also cause natural hazards where none existed before. Volcanoes erupt
periodically, but it is not until the rich soils formed on their eject are occupied by farms and human
settlements that they are considered hazardous.
Human intervention reduces the mitigating effect of natural ecosystems. Destruction of coral reefs,
which removes the shore's first line of defense against ocean currents and storm surges, is a clear
example of an intervention that diminishes the ability of an ecosystem to protect itself. An extreme
case of destructive human intervention into an ecosystem is desertification, which, by its very
definition, is a human-induced "natural" hazard. Quasi-natural and na-tech are terms used to denote
such hybrids.
If human activities can cause or aggravate the destructive effects of natural phenomena, they can also
eliminate or reduce them.
Controllable events vs. immutable events
For some types of hazards the actual dimensions of the occurrence may be altered if appropriate
measures are taken. For others, no known technology can effectively alter the occurrence itself. For
example, construction of levees on both sides of a stream can reduce the extent of inundations, but
nothing can moderate the ground shaking produced by an earthquake.
27.6.More Hazard Terminology
Secondary hazards
These are hazards that follow as a result of other hazard events. Hazards secondary to an earthquake
may be listed as follows to illustrate the concept. Primary hazard is the earthquake. Secondary hazards
are
 Building collapse
 Dam failure
 Fire
 Hazardous material spill
 Interruption of power/ water supply/ communication/ transportation/ waste disposal
 Landslide
 Soil liquefaction
 Tsunami (tidal wave)
 Water pollution
123
Rate of onset
Include rapid-onset and slower-acting (slow onset) natural hazards.
The speed of onset of a hazard is an important variable since it conditions warning time. At one ex-
treme, earthquake, landslides, and flash floods give virtually no warning. Less extreme are tsunamis,
which typically have warning periods of minutes or hours, and hurricanes and floods, where the likeli-
hood of occurrence is known for several hours or days in advance. Volcanoes can erupt suddenly and
surprisingly, but usually give indications of an eruption weeks or months in advance. (Colombia's Vol-
cán Ruiz gave warnings for more than a year before its destructive eruption in 1985.)
Other hazards such as drought, desertification, and subsidence act slowly over a period of months or
years. Hazards such as erosion/sedimentation have varying lead times: damage may occur suddenly as
the result of a storm or may develop over many years.
27.7.Hazard Identification and Assessment
If you are living in an area exposed to multiple hazards, for each hazard, ask yourself the following
questions and try to answer them. The answers are usually based on past experience of hazard events.
They may be recorded or may be gathered through interviews. They could give you a reasonable indi-
cation of the threat posed by the hazard for the area you live in.
 Could this hazard affect the area you live in?
 Is this hazard a significant threat there?
 How often does it pose a threat? E.g. Once every 5 years? 10 years?
 What is a close estimate of the population size that could be affected by this hazard event? Give a
rating. Very high? High? Medium? Low?
 What is the expected duration of the hazard?
 What is the expected damage from the hazard event? Give a rating. Very high? High? Medium?
Low?
 What is the expected intensity of impact expected? Give a rating.
 Very high? High? Medium? Low?
 How predictable is the threat?
 Can the effect of the event be reduced?
Hazard Assessment is sometimes called Hazard Evaluation or Hazard Analysis (UNDRO, 1991).
There seems to be a lack of consistency in the use of this terminology.
Hazard Assessment is the process of estimating, for defined areas, the probabilities of the occurrence
of potentially-damaging phenomenon of given magnitude within a specified period of time.
UNDRO
Governments mostly carry out hazard reduction measures without much enthusiasm and within eco-
nomic constraints. In order to compete for limited expenditure and resources hazard reduction propos-
als must find justification. This is facilitated by the collection of information about
124
 Local hazards - Location & Probability

 The extent to which they threaten local populations - Severity
 Ease with which their effects can be averted - Manageability
The severity of a natural hazard is quantified in terms of the magnitude of occurrence, which is an
event parameter. It can also be done in terms of the effect of the occurrence at a particular location.
This is called a site parameter. Both parameters may be combined in certain situations. Parameters for
selected hazards are listed below.
Table 31 Event and site parameters of selected Hazards

Natural Hazard Event Parameter Site Parameter
Cyclone Wind speed - km/h Area affected
Earthquake Magnitude – Richter Scale Intensity –Modified Mercalli Scale
Flood Area flooded – Depth of flood water - meters
km2Volume of water –
m3Speed
Landslide Volume of material Ground displacement - meters
dislodged, Area affected
Tsunami Height of wave crest Depth of flood water
Volcano Eruption size and duration Ash fall – meterLava flow - area
Hazard assessment approaches
The process of collecting this information is called hazard assessment.

These studies rely heavily on
 Available scientific information, including geologic, geomorphic, and soil maps; climate and
hydrological data; and topographic maps, aerial photographs, and satellite imagery.
 Historical information, both written reports and oral accounts from long-term residents. These
may include myths and legends.
For assessment of most natural phenomenon, one cannot expect complete data required to carry out a
comprehensive assessment. Depending on the situation, various methods are used with obvious
variations in the degree of accuracy.
Quantitative approach
Here mathematical functions are used to denote relationships between variable considered to quantify
the hazard. Numerical data can be fed in to assess the impact of the hazard event. An example is the
probable flood that a particular rainfall could cause within a watershed area. Flood dimensions such as
depth of flood and area of inundation would depend on the volume of water that flows into the stream.
Surface run-off, soil permeability, vegetation cover etc would determine this. The empirical data col-
lected from historical records as well as theoretical data from basic principles of physics are used to de-
rive the relationship between variables. The mathematical expression so derived could be used to fore-
125
cast future events. However, quantitative assessment may not be possible for all hazard events.
Qualitative approach
This method uses ranking such as ‘high’, ‘moderate’ and ‘low ‘to assess a hazard event. Where there is
a lack of sufficient data for quantitative evaluation, or where certain variables cannot be expressed
numerically, this qualitative ranking may be appropriate to take hazard mitigation decisions.
Probabilistic approach
After identifying the hazards that affect the planning area and assessment of the impacts from those
hazards, a probability analysis is undertaken. It provides an estimate of the probability of each hazard
affecting an area or region.
Probability for each hazard may be categorized as ‘high’, ‘moderate’ or ‘low’.
Probability of occurrence can be calculated through research on past events.
The Outcome
The outcome is natural hazards information, which denotes the presence and effect of natural phenom-
ena. Hazard assessment is the first step for hazard mitigation planning. It prioritizes hazards so that a
community or a government may use discretion to plan and implement hazard mitigation action.
This information should ideally include the location, severity, frequency, and probability of occurrence
of a hazardous event. Location is the easiest for planners to find; the rest can often be obtained from
sectoral agencies, natural hazard research and monitoring centers and integrated development planning
studies.
It could have information on natural ecosystems (e.g., slopes and slope stability, river flow capacity,
vegetation cover), which provides the basis for estimating the effect natural hazards can have on these
systems. Change in the ecosystem may create, modify, accelerate, and/or retard the occurrence of a
natural event.
Large-scale data describing lifeline infrastructure and human settlements for example, are critical
elements for preparing vulnerability assessments and for initiating disaster preparedness and response
activities.
Hazard Mapping
This is the process of establishing geographically where and to what extent particular phenomenon is
likely to pose a threat to people, property, infrastructure and economic activities. -UNDRO
126
Probability of hazard occurrence varies from place to place. The use of mapping to synthesize data on
natural hazards and to combine these with socioeconomic data facilitates analysis. It improves
communications among participants in the hazard management process and between planners and
decision-makers. Two important techniques in use are
 Multiple hazard mapping and
 Critical facilities mapping
Multiple Hazard Mapping (MHM)
This is usually carried out with new development in mind. Valuable information on individual natural
hazards in a study area may appear on maps with varying scales, coverage, and detail, but these maps
are difficult to use in risk analyses due to the inability to conveniently overlay them on each other for
study. Information from several of them can be combined in a single map to give a composite picture
of the magnitude, frequency, and area of effect of all the natural hazards.
 Regional scale hazard mapping uses 1:100,000 to 1:250,000. These are useful during planning
stages of regional development.
 Urban land use planners may need medium scale hazard maps of 1: 10,000 to 1:25,000.
 Site investigation for infrastructure projects may require large-scale hazard maps of 1:1,000 to
1:5,000.
 The multiple hazard map (MHM; also called a composite, synthesis, or overlay map) is an excel-
lent tool for fomenting an awareness of natural hazards and for analyzing vulnerability and risk,
especially when combined with the mapping of critical facilities. Its benefits include the follow-
ing:
 Characteristics of the natural phenomena and their possible impacts can be synthesized from
different sources and placed on a single map.
 It can call attention to hazards that may trigger others (as earthquakes or volcanic eruptions trigger
landslides) or exacerbate their effects.
 A more precise view of the effects of natural phenomena on a particular area can be obtained.
Common mitigation techniques can be recommended for the same portion of the study area.
 Sub-areas requiring more information, additional assessments, or specific hazard-reduction
techniques can be identified.
 Land-use decisions can be based on all hazard considerations simultaneously.
The use of a multiple hazard map also has several implications in emergency preparedness planning:
 It provides a more equitable basis for allocating disaster-planning funds.
 It stimulates the use of more efficient, integrated emergency preparedness response and recovery
procedures.
 It promotes the creation of cooperative agreements to involve all relevant agencies and interested
groups.
The base map upon which to place all the information is the first consideration. It is usually selected
during the preliminary mission. If at all possible, it is best to use an existing map or controlled photo-
graph rather than go through the difficult and time-consuming process of creating a base map from
scratch.
127
The scale used for an MHM depends on the hazard information to be shown, availability of funds and
the scale of the base map. If a choice of scales is available, then the following factors should be consid-
ered:
 Number of hazards to be shown.
 Hazard elements to be shown
 Range of relative severity of hazards to be shown.
 Area to be covered.
 Proposed uses of the map.
Much hazard information will be in forms other than maps, and not readily understandable by laymen.
It must be "translated" for planners and decision-makers and placed on maps. The information should
explain how a hazard may adversely affect life, property, or socioeconomic activities, and must there-
fore include location, likelihood of occurrence (return period), and severity. If some of this information
is missing, the planning team must decide whether it is feasible to fill the gaps.
Development and investment decisions made in the absence of these data should be noted. Despite the
importance of multiple hazard maps in the integrated development planning process, planners and de-
cision-makers must remember that the credibility, accuracy, and content of an MHM are no better than
the individual hazard information from which it was compiled. Furthermore, since it contains no new
information - it is merely a clearer presentation of information previously compiled - the clarity and
simplicity of the map is the key to its utility.
Critical Facilities Mapping (CFM)
This is carried out for development within existing infrastructure in mind. The term "critical facilities"
means all man-made structures or other improvements whose function, size, service area, or unique-
ness gives them the potential to cause serious bodily harm, extensive property damage, or disruption of
vital socioeconomic activities if they are destroyed or damaged or if their services are repeatedly inter-
rupted.
The primary purpose of a critical facilities map (CFM) is to convey clearly and accurately to planners
and decision-makers the location, capacity, and service area of critical facilities. An extensive number
of such facilities can be presented at the same time. Also, when combined with a multiple hazard map,
a CFM can show which areas require more information, which ones require different hazard reduction
techniques, and which need immediate attention when a hazardous event occurs. Some of the benefits
of a CFM are:
 The uniqueness of service of facilities in the area (or lack of it) is made clear.
 Facilities that may require upgrading and expansion are identified.
 The impact of potential development on existing infrastructure can be assessed before a project is
implemented.
 Any need for more (or better) hazard assessment becomes apparent
128
Combining critical facilities maps and multiple hazard maps
There are many advantages in combining a CFM, with a MHM, and integrating both into the develop-
ment planning process. For example, if a critical facility is found to be in a hazardous area, planners
and decision-makers are alerted to the fact that in the future it may confront serious problems. Its
equipment, use and condition can then be analyzed to evaluate its vulnerability.
If appropriate techniques to reduce any vulnerability are incorporated into each stage of the planning
process, social and economic disasters can be avoided or substantially lessened. Avoiding hazardous
areas, designing for resistance, or operating with minimal exposure, can make new critical facilities
less vulnerable.
Mitigation strategies for existing critical facilities include relocation, strengthening, retrofitting, adding
redundancy, revising operations, and adopting emergency preparedness, response, and recovery pro-
grams.
The benefits obtained by combining a CFM and an MHM include:

 Project planners and decision-makers are made aware of hazards to existing and proposed critical
facilities prior to project implementation.
 The extent to which new development can be affected by the failure or disruption of existing
critical facilities as a consequence of a natural event can be determined.
 More realistic benefit-cost ratios for new development are possible.
 Sub-areas requiring different assessments, emergency preparedness, immediate recovery, or
specific vulnerability reduction techniques can be identified.
Mapping techniques and tools
Community knowledge: - A simple mapping of local experience can be achieved using local knowl-
edge. Tools used in rural development activities such as
Participatory Rural Appraisal (PRA): can be very useful in this work. The method is cost effective and
the outcome reflects the local perception of hazard. The information can overlay local contour maps.
Surveys on historic events: There may be reports compiled on historic events, which may focus on
varying issues depending on its original purpose. However they may contain useful information.
Scientific investigation and research: Usually carried out through teamwork with experts from an array
of different disciplines. For example, landslide hazard mapping would require skills of geologists, geo-
technical engineers, geomorphologists, topography and so on.
Data over large areas for extensive time periods are collected. These are multidisciplinary studies and
each discipline would provide tools and techniques, which become more sophisticated and more accu-
rate over time.
129
Computer modeling using such data has opened up new vistas for hazard prediction. Geographic Infor-
mation Systems (GIS) modeling is one outstanding example. Remote sensing by satellite (RS) refers to
the viewing of the earth’s surface using sensing devices fixed onto satellites in orbit. Such data have al-
ready proved useful in flood prediction in Bangladesh for example. The future holds promising
prospects for this area of study.
28.TECHNICAL ANNEX 17: TRAINING – VULNERABILITY AND RISK
Goal (Session 2)
To instill an understanding of the concepts of vulnerability and risk and to develop the capability for
risk assessment.
Learning Objective
 The participant will be able to perform a vulnerability and risk assessment for a selected
community
Learning Outcomes
 The participant will learn to
 Distinguish between the concepts of vulnerability and risk
 List and describe criteria that add to the vulnerability of a community
 List and describe criteria that reduce the vulnerability
 Name and explain the components in quantification of risk
 Give a breakdown of elements at risk
28.1.Vulnerability
Vulnerability definition
A set of prevailing and consequential conditions – physical, social, and attitudinal – which adversely
affect the community’s ability to prevent, mitigate, prepare and respond to the impact of a hazard
event. The predisposition to suffer damage due to external events.
This definition is more focused on communities - Vulnerability is a condition or a predisposition. It ap-

plies to individuals, groups of individuals or communities, but it can be also used when referring to
physical structures or the environment in general. Vulnerability is about Susceptibility and Resilience
under threat of a hazard event.
Susceptibility: Proximity and exposure to an event. It is the potential to incur harm or avoid loss. It is
the fact of being exposed. You can be susceptible but not vulnerable. E.g. a landslide is threatening a
130
house but the owners have built a wall to protect it and to divert the landslide. Susceptibility is easy to
assess.
Resilience: Access to resources and capacities which determine the ability to recover from the impacts
of to a hazard event It is the ability to adjust and recover. E.g. the owners of the house threatened by a
landslide have a second house in town. They reside there during the rainy season. One can be
susceptible, but if he/she is resilient, one is not vulnerable. Resilience has many components: It implies
access to resources, individual skills, beliefs, etc. Compared to susceptibility, it is more difficult to
assess. E.g. Most Middle Eastern countries are in deserts. But their water supply system helps them not
to be exposed to drought conditions in normal life.
If susceptibility is very low and resilience very high, then one has minimum vulnerability. E.g. take a
displaced population in an emergency settlement. Susceptibility to measles is very high. If all children
are immunized however, resilience is high, and the vulnerability would be low. When Susceptibility is
high and resilience very low, one has maximum vulnerability. If the children are not immunized,
resilience is very low and the vulnerability is high. (Source: the Australian Emergency Management
Society).
Exposure
It is the state of being physically affected from a hazard. Researchers differentiate between voluntary
and involuntary exposure to hazards. Examples of involuntary exposure include air pollution (as we
must breathe ambient air), toxic contamination of food (as we must eat), and water pollution (as we
have to drink). We do, on the other hand, have a greater choice over where we live and what activities
we engage in (living in coastal or seismically active zones is to some extent voluntary; smoking and
exposing yourself to the threat of cancer is definitely voluntary).
Capacity
Those positive conditions or recourses which increase the ability of a community to deal with hazards.
Capacity may be
 Physical
 Social/Organizational
 Attitudinal/Motivational.
Capacity is also reflected in the preparedness of the community to face a hazard event.
Preparedness
Measures taken in anticipation of a disaster to ensure that appropriate and effective actions are taken in
the aftermath of a hazard event. To enhance preparedness, people plan how to respond in case a hazard
event occurs and work to increase the resources available to respond effectively. Preparedness
activities are designed to help save lives and minimize damage by preparing people to respond
131
appropriately.
Response
Refer to actions taken immediately following the impact of a hazard event when exceptional measures
are required to meet the basic needs of the survivors. –ADPC. It refers to the sum of all actions taken
to adjust to hazards; more narrowly defined to mean the appropriate actions taken during an emergency
to protect people and the things they value from harm, rescue them, and facilitate the transition to post-
disaster recovery.
Socio-economic indicators
Socio-economic indicators provide another dimension to view Vulnerability. The Indicator of Human
Development (IDH) is synthesized by combining other indicators and gathers together
 The adjusted GDP per capita
 Life expectancy
 Adult literacy.
The assumption is that the lower the IDH, the lower will be the mean wealth, the literacy and the aver-
age health state of the population. This will increase the vulnerability to physical hazards. Poverty is
one of the major vulnerability criteria. Poverty also has an effect on housing which constitute a usually
high damage percentage in case of disaster (Center for Research on the Epidemiology of Disasters,
CRED).
Demographic indicators
When high to very high population densities (>200 hab/km2) are combined with unfavorable socio-
economic parameters (low IDH, high birth and mortality rates), the vulnerability is expected to be high
(CRED).
28.2. Risk
Risk definition
Risk is the likelihood or probability of a hazard event of a certain magnitude occurring. Risks are
measures of the threat of hazards. Risk is the actual exposure of something of human value to a hazard.
Often regarded as the product of probability and loss. Risk is the exposure or the chance of loss due to
a particular hazard for a given area and reference period. It may be expressed mathematically as the
probability that a hazard impact will occur multiplied by the consequences of that impact.
(Note: Definitions of risk in the hazards literature vary from those that equate risk with probability to
those that see risk as the product of a probability and a particular kind of impact occurring.)
132
Risk and Hazard are two concepts that are different from each other. The ocean is a hazard (deep water
and large waves). If one attempts to cross the ocean in a small rowboat, a great risk (probability of cap-
sizing and drowning) is incurred. If the crossing is made aboard the Queen Elizabeth, (a large passen-
ger ship) the risk is reduced – all else being equal. The ocean going vessel is a device used as a safe-
guard against the hazard. In general, risk may be diminished by increasing safeguards but never elimi-
nated unless the hazard itself is removed.
Consequence: The monetary and non-monetary "costs" or “losses” of a hazard event. This includes
financial, economic, life safety, environmental, social, legal and other costs or losses. Its assessment
may be made easier if the elements at risk are enumerated first.
Elements at risk: Persons, buildings, property, crops, utilities, critical facilities, infrastructure, environ-
ment or societal components with a potential of being exposed to a hazard event and likely to be ad-
versely affected by the hazard event.
Elements at risk may be listed as follows:

Table 32 Elements at risk
Physical Infrastructure, Roads, Railway, Bridges, Harbor, Airport,
Critical facilities Emergency shelters, Schools, Hospitals and Nursing Homes, Fire Brigades,
Police,
Utilities Power supply, Water supply, Transport, Communication, Government services
Economic Business and trade activities, Access to work, Impact on work force, Opportunity
cost
Societal Vulnerable age categories, Low income group people, Gender
Environmental Loss of biodiversity, Damaged landscape, Physical and chemical changes in the
surroundings
Quantification of risk: There are three essential components to the quantification of risk after
identification of local hazards.
Table 33 Parameters for quantifying risk

Hazard occurrence probability (p) the probability of occurrence of a specified natural hazard at a
specified severity level in a specified future time period.
Elements at risk an inventory of those people or things, which are exposed to the
hazard.
Expected loss (L) the degree of loss to each element, should a hazard of a given
severity occur. Its accuracy and acceptability will depend on the
methodology used to derive it.
Disaster losses include the direct impacts like the loss of life, housing and infrastructure as well as
indirect impacts on production in utility services, transport, labor supplies, suppliers and markets.
Secondary losses include impacts on macroeconomic variables such as economic growth, balance of
payments, public spending and inflation. The impacts are felt more by developing countries.
133
In 1977, a cyclone struck the Andrah Pradesh. The following loss figures were estimated for the East
Godavari District – Houses destroyed 2,89,906, Houses damaged 89,677, TOTAL 3,79,583, Persons
rendered homeless 1.442 million.
If measures were taken to strengthen houses by retrofitting, the loss would have been reduced as
follows –Houses destroyed Nil, Houses damaged 1,36,489, Persons rendered homeless 0.519 million.
The economic benefit is estimated to be Rs. 91 crores.
It is advisable that, if a hazard event does occur, that a post-audit be carried out so that one may match
the ‘actual’ Vs the ‘expected’. This would give more understanding on deficiencies of the Risk Assess-
ment carried out and help to improve the process next time around.
Note: The probability of occurrence of natural hazards events may be estimated by statistical extrapo-
lation from historical data. The accuracy of such estimates depends on the completeness of data and the
period of time over which it has been collected.
Losses are measured differently for different hazards, by different agencies, and by different users.
Most loss data are dollar estimates, but some dollar estimates are for specific agencies or only one lev-
el of government. The estimates could show variability as well as considerable uncertainties and is
valid for only a short period of time.
Risk assessment
It is the overall process of identifying and analyzing risks. The process of characterizing hazards
within risk areas, analyzing them for their potential mishap consequences and probabilities of
occurrence, and combining the two estimates to reach a risk rating.
Experts on behalf of others carry out assessment of the risk parameters.

Risk Assessment provides a sound basis for mitigation planning and for allocation of funds and other
resources.
Some use the term Hazard Vulnerability Analysis (HVA) to express Risk.
It is the process of evaluating risk associated with a specific hazard and defined in terms of probability
and frequency of occurrence, magnitude and severity, exposure and consequences.
Frequency vs. Severity: Where flooding occurs every year or every few years, the hazard becomes part
of the landscape, and projects are sited and designed with this constraint in mind. Conversely, in an
area where a tsunami may strike any time in the next 50 or 100 years, it is difficult to stimulate interest
in vulnerability reduction measures even though the damage may be catastrophic. With so long a time
horizon, investment in capital-intensive measures may not be economically viable. Rare or low-proba-
bility events of great severity are the most difficult to mitigate, and vulnerability reduction may de-
mand risk-aversion measures beyond those justified by economic analysis.
Here, severity means the event’s duration and impact area.
134
There are many methodologies for Risk Assessment from High Tech computer-based methods to pen-
cil pushing and head scratching. Computer-based Geographic Information Systems (GIS) modeling use
many different kinds of information to assess risk. However it needs sophisticated hardware, software
and expertise of handling them.
Risk matrix: A simpler method is the Risk Matrix Analysis. The method gives a qualitative measure
that permits the prioritization of risk among multiple hazards. It enables hazard mitigation planners to
classify various types of hazards into different categories of priority by locating them on a two-
dimensional grid based on their probability and loss. The ranking of ‘high’, ‘moderate’ and ‘low’ is
subjective and would vary from one group to another. The ranking depends on Probability of a hazard
event and Potential loss.
Figure 43 Diagram of risk matrix analysis
135
Table 34 Guidelines to do a matrix analysis

Probability High Events that occur more
frequently than once in 10 years
Moderate Events that occur from once in
10 years to once in 100 years
Low Events that occur from once in
100 years to once in 1000 years
Very Low Events that occur less frequently
than once in 1000 years
Loss Based on the potential by taking Fatalities
into account elements at risk Injuries
Impact on facilities, critical
services and infrastructure
Property damage
Business interruption
Environmental/Economic
impact
The criteria used for class categorization is also subjective. What has been given below merely illus-
trates the type of criteria that countries may adopt to facilitate the process.
Table 35 Risk categorization by class

Class A High-risk condi- Immediate action Possible deaths People affected
tion is necessary over 1000 may be over
100,000 Complete
shut down of facil-
ities and critical
services for more
than 14 days. Over
50% of property
located in the area
may be damaged
Class B Moderate to high- Prompt attention Possible deaths People affected
risk condition needed less than 1000 may be between
where risk can be 50,000 to 100,000
reduced by mitiga- Complete shut
tion activities and down of facilities
contingency plan- and critical ser-
ning vices for 7
days25% of the
property located in
the area may be
damaged
Class C Low risk condi- Possibility of People affected
tion However in- death low between 10,000
136
vestment in fur- and 50,000Com-

ther mitigation plete shut down of
and planning may facilities not more
be necessary after than 1 day. About
cost-benefit analy- 10% of the proper-
sis ty located in the
area may be dam-
aged.
Class D Very low risk No possibility of People affected
condition Only death less than 10,000
limited action Facilities and
necessary critical services
may not be
affected. About
1% of the property
located in the area
may be damaged
Risk as a function of hazard, loss and preparedness. According to Fournier d’Albe (1979), risk may be
conceptualized as follows:
Risk = Hazard (probability) x Loss (expected)
------------------------------------------------------
Preparedness (loss mitigation)
 Physical/material
 Social/organizational
 Attitudinal/motivational aspects.
Thus their quantification cannot be discrete. It is advisable therefore, not to treat this conceptualization
as a mathematical entity because of the emotive aspects that are inclusive in “preparedness”. It is wiser
to consider that disaster risk is a function of preparedness/capacity/manageability.
Greater the hazard probability, greater the risk.

Greater the loss, greater the risk.
Greater the preparedness or capacity of a community, lesser the risk.
(Capacity and Risk show an inversely proportional relationship)
Risk = function (hazard, loss, preparedness)
Preparedness has an inverse relationship to risk.
137
29.TECHNICAL ANNEX 18: TRAINING – DISASTER MANAGEMENT
Goal (Session 3)
To instill an understanding of the disaster management process
Learning outcomes
After completing this session, the participant will be able to --
 Affirm the usefulness of integrating management principles in disaster mitigation work
 Distinguish between the different approaches needed to manage pre- during and post- disaster
periods
 Explain the process of risk management
Learning objectives
Participant will learn to --
 Distinguish between disaster management and risk management
 Explain the disaster management Cycle
 Describe the strategies for risk mitigation
 List activities needed for post-disaster management
What is Management?
Management consists of decision-making activities undertaken by one or more individuals to direct

and coordinate the activities of other people in order to achieve results, which could not be accom-
plished by any one person acting alone.
Management is required when two or more persons combine their efforts and resources to accomplish
a goal, which neither can accomplish alone.
29.1. Disaster Management
What is disaster management?
Disaster management can be defined as the effective organization, direction and utilization of
available counter-disaster resources.
Disaster management includes administrative decisions and operational activities that involve –
Prevention, Mitigation, Preparedness, Response, Recovery and Rehabilitation.
Disaster management involves all levels of government. Non-governmental and community-based

organizations play a vital role in the process.
Modern disaster management goes beyond post-disaster assistance. It now includes pre-disaster plan-
ning and preparedness activities, organizational planning, training, information management, public re-
lations and many other fields. Crisis management is important but is only a part of the responsibility of
138
a disaster manager. The newer paradigm is the Total Risk Management (TRM) which takes a holistic
approach to risk reduction.
The traditional approach was to provide immediate humanitarian aid (usually rescue teams, materials
and medical services) as quickly as possible after the onset of a disaster. There has been a paradigm
shift over the last decade.
The modern view is that there must be pre-disaster mitigation measures to avoid or reduce impact of
disasters. Pre-disaster measures to prevent or mitigate disasters are called Risk Management.
Disaster management cycle
The traditional approach to disaster management has been to regard it as a number of phased sequences
of action or a continuum. These can be represented as a cycle.
Figure 44 Diagram of the Disaster Management Cycle
In this model, disaster management occurs in stages, in sequence. The focus is more on activities
immediately before and after the onset of the disaster event.
Mitigation and preparedness precede a disaster. Pre-disaster management is called Risk Management.
29.2. Risk Management
The process, by which assessed risks are mitigated, minimized or controlled through engineering,
management or operational means. This involves the optimal allocation of available resources in
support of group goals.
139
Risk identification has already been discussed. It is usual to allocate risk management to a special body
at national level. Usually it is a National Disaster Management Organization (NDMO). At local level it
may be the responsibility of a Disaster Mitigation Committee, which administers risk management.
This varies in different countries depending on administrative patterns and needs.
Risk reduction
Effective risk reduction involves mitigation measures in hazard prone areas. It may also involve
overcoming the socioeconomic, institutional and political barriers to the adoption of effective risk
reduction strategies and measures in developing countries.
Mitigation
Measures taken prior to the impact of a hazard event to minimize its effects (which may be structural
and non-structural). The following non-structural and structural means may be listed (Carter, 1991)
Non-structural mitigation
Legal framework - Examples are building codes for built structures to withstand impact such as
cyclones or earthquakes etc.
Land-use planning - Controlling human activities in hazard prone areas (zoning) to avoid fatalities and
loss. This may involve re-location of communities to safer locations. It may be achieved by passing
legal statues, E.g. Ordinances.
Incentives - Often provide better inducements for mitigation than legal impositions. Government grants
or subsidies may help to persuade commercial and other institutions to include mitigation measures in
their building and reconstruction. Insurance companies may be persuaded to offer reduced premiums
for buildings that incorporate hazard resistant measures.
Training and Education - Provide awareness and know-how to those government officials involved in
disaster management, construction experts, craftsmen, land use planners and the general public.
Public Awareness - This is necessary to ensure: a good public knowledge and understanding of natural
hazards and vulnerabilities, awareness of effective mitigation measures, public participation in
community preparedness programs.
Institution building - This is the strengthening of national or community social structure. This can work
through; Identifying and strengthening organizations that serve as coping mechanisms: by increasing
capacity and skills to face a crisis. Increasing the number of coping mechanisms within a country or
community and by linking them to outside resources and Encouraging actions that promote
cooperation among different groups within society.
Structural mitigation
140
Engineered structures - These involve architects and engineers during the planning, designing and
construction phases. The application of sound technical principles is achieved through
 Site planning
 Assessment of forces created by natural hazards
 Planning and analysis of structural measures to resist such forces
 Design and proper detailing of structural components
 Construction with suitable material
 Good workmanship under adequate supervision
Most countries have building codes for engineered construction.
Non engineered structures - These are constructions by owners using local masons and carpenters who
lack formal training. The design may be improved according to traditional ways. Their location on
hazard prone areas may be controlled.
The mitigation plan
The disaster mitigation committee should ‘brainstorm’ on all possible measures that might help to re-
duce risk. The alternatives should be weighed and the more acceptable ones selected which are appro-
priate to satisfy community needs. Then a plan must be formulated to facilitate the implementation of
the selected risk reduction measures.
29.3. Prevention
Prevention is more applicable to man-made and technological disasters. Stringent safety precautions
through technological innovation can bring these about. Natural hazard events cannot be prevented.
But if the vulnerability of the community is reduced, one can prevent the hazard event becoming a
disaster. The government departments or municipalities can prevent disaster impact by conducting
selected mitigation activities before a disaster strikes.
A dam could control floodwater. Controlled burning of fire belts could prevent the spread of a wild
fire. Stringent building code imposition can reduce collapse of buildings during an earthquake. Proper
socio-economic development and active ownership and participation of communities in the disaster
management continuum and the development of adequate warning systems where applicable, can also
bring positive results.
29.4.Crisis Management
Crisis management applies to emergency operations and covers Preparedness and Immediate post-
disaster period. To avoid confusion during the emergency period immediately after the disaster strikes,
disaster management places heavy emphasis on advance planning. This is called disaster preparedness.
141
A variety of different management systems have evolved to respond to different types of disasters.
Most agencies borrowed military and or business organizational models to mange disasters. These used
a pyramidal hierarchy of upper-level managers, middle managers and field managers. The upper-level
and middle managers dealt with managing the organization and facilitating field operations. The field
manager was responsible for development of programs that directly assisted people of the disaster im-
pact area.
Recently, new management models have appeared which emphasize community participation in deci-
sion-making and response.
29.5.Response and Relief
If a disaster occurs, response and relief have to take place immediately. Rescue of affected people,
distribution of basic supplies such as food water, clothing, shelter and medical care become urgent
need of the hour. Delays will occur if government departments and municipalities have no clear plans
to manage such events. It is therefore important to have plans in place.
Take a simple example. A flood has occurred in a mountainous area and there is very strong wind and
continuous heavy rain. The possibility of landslides is real. Members of the public are panicking and
the mayor is under pressure to take emergency action
A well-managed team of government and local players should be prepared and know where to go,
what to do. If the situation is managed in an ad-hoc way, the affected people will rush off in all direc-
tions, waste valuable time, and even make serious mistakes with fatal consequences.
Search and rescue plans need to be clear and all role players need to know their role and functions in
such activities. Basic needs such as emergency shelter, water, food, and medical care have to be pro-
vided. A plan must be in place.
29.6.Rehabilitation and Reconstruction
The rehabilitation period involves the weeks and months after the disaster. The focus is to enable the
area to start functioning again. This involves debris removal, restoration of public services and
provision of temporary housing. Reconstruction is a much longer-term activity. This phase involves
permanent rebuilding, improved infrastructure and better disaster planning. Both rehabilitation and
reconstruction phases demand good management. Diversion of national and international aid
prudently, prioritization of activities, proper coordination and monitoring as well as prevention of
corruption and abuse of scarce funds become priorities.
142
30.TECHNICAL ANNEX 19: TRAINING – PARTICIPATORY RURAL APPRAISAL TOOLS
Goal (Session 4)
To impart knowledge and skill to use tools for participatory appraisal of hazards, vulnerabilities and
risks within a community.
Learning Outcomes
 After completing this session, the participant will be able to perform a participatory appraisal of
hazards, vulnerabilities and risks for a selected community
Learning Objectives
 Participant will learn to

 Describe the different tools used in participatory appraisal
 Conduct participatory assessment of hazards, vulnerabilities and risks
Content of this session is based on a Hand Book produced by the National Institute of Rural
Development (NIRD), Hyderabad, India.
This session will outline tools that have been recommended for working with rural communities.
However, PRA tools can be used in almost any situation and need not be confined to appraisal
interviews in rural situations.
The major challenges in vulnerability reduction in communities are: sensitization to bring about
attitudinal and behavioral change; using participatory techniques to build rapport, elicit support,
information and participation of the people in their own development.
30.1.Importance of Participatory Techniques
Participatory techniques aim to ‘break the silence’ of the poor and disadvantaged sections, recognize
the value of popular collective knowledge and wisdom and legitimize the production of knowledge by
the people themselves. Participatory approaches seek to be catalysts enabling and empowering the
people. These have internalized some key techniques in adult training for learning such as
 linking learning to problems,
 linking learning to people’s goals and visions,
 giving trainees control over decisions on training.
The participatory approach emphasizes flexible learning, is adaptable to the pace set by the
learners/trainers and tailored to needs expressed by participants themselves.
143
30.2.Participatory Rural Appraisal
Participatory rural appraisal (PRA) is a methodology to enhance the development agent’s

understanding of the rural reality for the planning and development of projects; and the feeling of a
greater degree of ownership and responsibility in the rural poor for better results and social acceptance
of the program. The effectiveness of participatory approaches has led donors, government
organizations and NGOs to use PRA in their programs. Participatory training is based on the belief that
learners with their life experiences are themselves a rich source of learning; learning cannot be
imposed; the learner can only be encouraged to learn; learners learn best by doing or practicing an
activity; and learning is facilitated by a positive/successful activity/experience resulting in further
achievement, thereby building up a ‘virtuous circle’.
30.3.Aims of participatory methods
Different participatory methods are used for different ends. Sometimes participatory approaches are
the means and ends as well. In the case of decentralized development, the ends are:
 peoples’ active participation in prioritizing needs/micro-planning;
 activating the key local institutions attitudinal and
 behavioral change in the bureaucracy.
Since village organizations have to play an active role in initiating the micro-planning exercise, they
need a locally relevant database that is validated by the local people. This will form the basis for
setting local priorities. This should help, in turn, in the formulation of local action plans in the form of
development activities/projects/programs.
30.4.Scope of PRA
PRA is used
 To ascertain needs
 To establish priorities for development activities
 Within the scope of feasibility studies
 During the implementation phase of projects
 Within the scope of monitoring and evaluation of projects
 For studies of specific topics
 For focusing formal surveys on essential aspects, and identifying conflicting group interests.
Areas of application
 Natural resource management
 Agriculture
 Poverty alleviation/women in development programs
 Health and nutrition
 Preliminary and primary education
 Village and district-level planning
144
 Institutional and policy analysis.
Table 36 Participatory rural appraisal

Principles and methods Benefits
From ‘they learn from us’ to . From ‘we let them participate’ to Empowering the poor and
‘we learn from them’. ‘they take command of their own pro- weak to assert their priorities,
cess’. make demands and act.
From ‘we’ve done a PRA’ From ‘we use instruments from our Expression and harnessing of
to ‘we admit being corrected toolbox’ to ‘they can map, model, esti- local diversity.
by people’. mate, score, analyze, plan themselves’.
From ‘we share our knowl- Offsetting biases: spatial, project, gen- Community participatory ap-
edge analysis with them’ to der/elite, seasonal calendar. praisal, planning, implementa-
‘we enable them to learn tion, monitoring and evalua-
from each other and conduct tion.
their own analysis’.
Rapid progressive learning, Triangulation: using different meth- Identification of research pri-
which is flexible, explorato- ods, sources and disciplines, and a orities; experts more receptive
ry, interactive and inventive. range of informants in a range of to the ability of rural poor to
places and cross checking to get closer design, implement and evalu-
to the truth through successive approx- ate.
imations.
Facilitation: to enable peo- Insights gained from PRA
ple to do more or all of the leading to policy change.
investigation themselves
and own the outcome.
Sharing: a culture of sharing Behavior and attitudes: critical self- A culture of open learning
information, methods, field awareness in external facilitators, among govt., NGOs and com-
experiences among NGOs, learning from errors. munity.
government and villagers.
Table 37 Tools of PRA

Diagram Priority ma- Seasonal Time trends Venn diagram
trix calendar
Map Transect Entire com- Helps to Provides local perspec- Helps to iden-
Provides alterna- Builds rapport munity in- identify tive on time changes in tify marginal-
tive database with locals volved in pri- lean peri- natural resources/ecolo- ized individu-
Depicts differing Supports maps oritizing ods for re- gy/etc. als and groups
local perception of local re- needs and de- sources within the vil-
of local prob- sources/needs velopment and timing lage
lems/needs initiatives of supply
of key
145
farm in-
puts
30.5.Participatory community monitoring and evaluation
Why participatory community monitoring and evaluation?
Participatory community monitoring and evaluation are extremely important for learning about the
achievement/deviation from original concerns and problems faced by local development
projects/programs being implemented, so that corrective measures can be taken in time.
Evaluation is often carried out by donor agencies or policy makers and helps in assessing whether the
project has brought benefits to those for whom it was intended. An evaluator is expected to examine:
whether it was right to have invested resources in the project in the context of competing needs;
whether the underlying assumptions and design were right; whether progress was made towards
planning changes, and if not, why; and unplanned changes that may have occurred.
Monitoring ensures that i) inputs are ready in time; ii) works plans are followed closely; iii)
adjustments can be made and corrective action taken as and when necessary; iv) people who need to
know are kept informed; v) constraints and bottlenecks are found; and vi) resources are used
efficiently.
Aim of participatory monitoring and evaluation (pme)
To assess information or generate data on development activities being carried out at the local
community level –
 To facilitate monitoring and evaluation by beneficiaries of different development activities.
 To increase beneficiaries’ commitment and understanding in designing, planning and
implementing community-based development projects or programs.
Participatory monitoring involves local beneficiaries in measuring, recording, collecting, processing

and communicating information to assist local development project extension workers and local group
members in decision-making.
Participatory evaluation assists in adjusting and redefining objectives, reorganizing institutional

arrangements or re-allocating resources as necessary. Monitoring and evaluation system (MES) allows
continuous surveillance in order to assess the local development project’s impact on intended
beneficiaries.
Involving local people in project evaluation is one of the learning objectives of participatory
management. Apart from project’s impact on the life of the people, it is also worthwhile to evaluate: i)
attitudinal changes in the local community about their role and sense of responsibility; ii) if people
have gained confidence in their ability to undertake new activities; and iii) lessons about people’s
146
capacity, extent of participation and community responsibilities.
It provides an opportunity to the project implementation committee to assess deficiencies in the project
design - if objectives and work plan were realistic, if local funding was adequate and whether project
actually owned by the people. Answers to these questions indicate future precautions and
modifications in the method and approach. This in itself is an achievement in capacity building at the
local level.
Role of community extension workers
It is the responsibility of extension workers/community development motivators to make beneficiaries

aware about the project/programs and their objectives. Extension workers should develop and help
beneficiaries identify indicators and measurements for each project activity. Based on these, extension
workers should collect data on inputs and outputs by using simple formats and tables. Extension
workers should process, organize and analyze the data for evaluation. For participatory evaluation,
they should assist beneficiaries to understand the process, using simple procedures. After processing,
organizing and analyzing the data, extension workers must assess the impact of local development
project activities.
PME should be:

 Demonstrative, not instructive in writing
 Collaborative, not individualist or directive
 Explorative, not repetitive
 Listening to, not lecturing
 Interactive, not dominating
 Qualitative, not quantitative
 For community/people, not project-oriented.
Table 38 Steps in participatory monitoring and evaluation (PME)

Step I Understanding goal/objectives of local development project/program.
Step II Identifying activities to achieve objectives.
Step III Identifying measurements to assess results or show extent of progress achieved.
Step IV Developing measurement indicators.
Step V Identifying methods and techniques of collecting information.
Step VI Selecting formats/visual tools for presenting information
147
31.TECHNICAL ANNEX 20: TRAINING – COMMUNITY PUBLIC AWARENESS

Goal (Session 5)
To impart an understanding of public awareness creation
Learning Outcome
 The participant will be able to explain the process of public awareness creation
Learning Objectives
 Explain situation analysis
 Describe audience analysis and audience segmentation
 Achieve SMART objective setting
 Understand the process of message creation
 Discriminate between available channels for message dissemination
 Evaluate effectiveness of public awareness activities
31.1.Definition of Social Change Communication Campaign
Some generic definitions for social change communication campaigns:
Programs designed to influence the voluntary behavior of target audiences where the primary objective
is to improve the welfare of the target audiences and/or the society of which they are part (Andreasen,
1995)
A target audience-based approach to promote socially beneficial behavior change in specific

populations
It is much more than education; it includes analysis, planning, execution, and evaluation of integrated
programs to influence people to trade their old ideas, beliefs, and behaviors for new ones.
You want to change people’s behaviors, but that’s not your only outcome. You’ve got to change the
environment.
As figuratively defined by someone, “An idea attached to leadership”. Very often, the “change” social
change communication campaigns seek is in knowledge, attitudes, skills and practice. A larger
definition is that community norms need to change so that the targeted knowledge, attitude and
practice changes become acceptable. Further, social change campaigns may also seek societal level
changes such as change in policy.
148
31.2.Definition of Disaster Risk Communication
Disaster Risk Communication is a specific application of social change communication campaigns to

bring about change in knowledge, attitude, skills and practice at the individual and possibly societal
level with regard to disasters.
Specifically, DRC is: “The systematic planning of information transfer, based on scientific research, to
prevent, solve or mitigate the risk problem with adjusted and customized information (risk messages)
for specific target groups” (Gutteling and Wiegman, 1996). The authors of this definition also point out
that the communication can be multi-way suggesting that it is not just a transfer of information. Let us
try to understand each part of this definition.
Systematic Planning
The term currently in use for communication campaigns is “public awareness.” Public awareness
draws on the mass media to create awareness in society about the development intervention, while
social marketing uses marketing ideas, i.e., the integrated use of channels, cost, intervention,
promotional ideas, etc., to promote and deliver social change interventions. What is relevant is that any
serious communication campaign, in this case for disaster risk communication, must be based on
systematic planning.
The generic communication model (Lasswell, 1948) is: Who? Says what? To whom? With what
effect?
It is critical that the “What” itself be determined based on feedback, so that both the content of the
message itself and the language/design, etc., used to convey the content is relevant and representative
of the target group, and clarity is assured (or in communication jargon noise is reduced). That is, rather
than “transfer” alone, the approach should be a mutual determination of the message and then its com-
munication to the larger target group.
The definition leaves out the “Who” or the source of the message. Sources of messages must be:
 Knowledgeable: Content experts must be used for technical message content in mediated
messages. While this might be routinely done for mass mediated messages, the need to impart
content knowledge to interpersonal communication sources, particularly local level implementing
officials, community opinion leaders, and volunteers is critical.
 Credible/Trustworthy: Whether it is a medium (radio, newspaper, television) or an interpersonal
source (government official), the source must be credible to the target and one that the target
group trusts.
The two characteristics are intertwined so that credibility may lead to trust and vice versa, but depend-
ing on the campaign sometimes it is credibility and at other times it is trustworthiness that is more im-
portant.
Also, while knowledge lends credibility, it is not the only determinant of credibility. Sometimes like-
ability may also determine credibility.
149
 Likeable: Likeability may be related to all the above concepts but is also related to attractiveness
of the source and to the next characteristic of a source, representative ness.
 Representative: A representative source is one who looks like, speaks like, and dresses like the
target audience. People are more able to relate to and understand sources that are similar to them.
Scientific Research
The importance of research, systematic, planned, scientific, is often overlooked. Research informs
each stage of the communication campaign and helps to ensure that decisions are made on the basis of
fact rather than whim or opinion alone. Often when campaigns may fail, the temptation is to blame the
ignorant public without looking at whether risk perception was accurate, whether the message was the
right one, presented in the right manner, for the right public, etc. Research assists with these decisions.
Problem definition: At this stage, there may be a disjoint in what the agency that is creating the
campaign perceives as the problem and what the target group sees as a problem. While we are
specifically interested in problems that can be solved by communication, we must be alert to the fact
that some non-communication problem may be the larger problem. It is of course assumed that before
we get to the stage of a communication campaign, it has been determined based on research that a
communication problem is present. Still taking the target audience’s views even in defining the
communication problem is important.
Research at this stage may be largely qualitative and secondary (i.e., collected from existing materials
rather than gathered from scratch) supplemented by quantitative and primary (i.e., gathered specifical-
ly for the purpose).
Target audience definition: At this stage, research must be undertaken to provide a complete profile of
the target audience starting with demographics. Such a profile is critical for the selection of messages,
tactics, channels, sources, and such.
Research at this stage is largely quantitative and primary but also qualitative and secondary.
Delineation of strategy: The main message strategy (i.e., the big message idea) must be based on
research into the problem and the target, but must be finalized based on testing with the target
audience. This may require both secondary and primary research, and both qualitative and quantitative
research.
Message creation: Three broad stages of message creation are concept, rough (sketch of the idea), and
comprehensive (close to final). At each of these stages, feedback from the target group is essential.
Essentially, this is pre-testing the message. Research at this stage is largely quantitative and primary
but also qualitative sometimes.
Channel determination: This is largely based on research into the target audience’s channel profile
(and knowledge of the strengths of each medium).
150
Evaluation: Evaluation is conducted before, during and after the campaign respectively to establish a
baseline, to see if the campaign is working and make adjustments if necessary, and to measure the
success of the campaign.
Largely, primary quantitative evaluation is done but qualitative evaluation is also used.
Prevent, Solve, Mitigate Risk Problem This is the main goal of the disaster risk communication
campaign.
Specific objectives will also need to be articulated as building blocks through the use of communica-
tion to achieve this larger goal.
Adjusted and Customized Risk Messages
The messages themselves are composed of the content as well as the language, design, etc., used in
constructing them. Messages need to be adjusted as noted above based on pre-campaign feedback and
concurrent campaign evaluation. They must also be customized to different target audiences or sub
groups within audiences, to different channels, and for different times of the year or day, etc.
Target Groups
As mentioned above, the target audience must be profiled in depth. Any campaign may be targeted at
more than one target audience (this does not imply that the same message, media, etc., be used for each
group).
These groups may be --
 Policy makers who, for example, may need to be convinced to make disaster preparedness a
priority or to reform building codes. This includes local, regional and central bodies that
determine and implement policies and practices.
 Community members (and social networks) who may need to change formal and informal social
norms or standards, for example, in terms of acceptance of education for females.
 Institutions and organizations that may assist in the campaign by becoming venues. These include
businesses, schools, religious institutions, public agencies, service organizations, trade or
professional associations.
 Interpersonal groups includes primary groups such as peers, family and friends that provide social
identity, support and role delineation for an individual and may be able to influence change in
individuals.
 Individuals who may change their cognitive, psychological, and behavioral makeup, i.e., their
knowledge, beliefs and attitudes, personality and action. Often in DRC campaigns, change at the
individual level is the major component.
151
31.3.Steps in DRC
The broad basic steps in DRC (all of which are informed by research) are (see more details in
appendix):
 Situation Analysis: Study the situation with regard to the problem, its history, past attempts to
resolve the problem, etc.
 Target Audience Definition: Determine your specific target audience(s) and profile it.
 Formulating Communication and Action Objectives: Outline objectives of the DRC in terms of
communication and action outcomes at several levels as appropriate from the individual to
societal.
 Formulating Message Strategy and Tactics: Determine the main message(s) and the different ways
of presenting the message.
 Formulating Channel Objectives, Strategy and Tactics: Determine the channel including
interpersonal sources through which the message(s) will be communicated.
 Producing Materials and Implementing Campaign
 Monitoring and Evaluation: Determine how to monitor the progress of the project and to evaluate
its success.
DRC is important because:
 It empowers people. Most DRC campaigns have the public as their target audience. By providing
the public with information, DRC campaigns empower them to control their lives and give them
strategies on how to cope when disasters strike. This empowerment can help reduce the fatalistic
attitude that one finds in many regions and ultimately save lives and livelihood, the most critical
resources of any country.
 DRC campaigns also enable policy change. This end result comes not only from the fact that
policy makers may be the target group for the campaign, but also from the fact that the
empowered public can create public opinion pressure to bring about change at the policy level.
 DRC campaigns also make the work of disaster relief a lot easier. If people have an accurate
perception of the risk to them, and know what to do and how to respond when the risk is present,
relief agencies will have less work because fewer lives will be at stake and more people will have
followed directions. That is, people will be better prepared and the work of agencies will be more
effective.
31.4.Barriers to Making Change
While the need for DRC is urgent, there are barriers to making change that DRC campaigns must keep
in mind:
 Negative demand: Target audiences who are being reached by such campaigns may not want it
(Andreason, 2000).
 High involvement issues: Many of the behaviors these campaigns try to influence are much more
involving, serious.
 Invisible benefits: Often times, the benefits of these campaigns are not immediate and obvious.
 Benefits to others: Often times, the benefits of these campaigns accrue to others.
152
Thus making social change is difficult, but the tools presented in this workshop will assist in the
endeavor. Apart from barriers to change, there are also factors that could become points of resistance.
These must be identified and dealt with. Resistance points may be:
 social
 cultural
 economic
 religious
 products of ignorance
31.5.Some Distinctions
“Communication for development” is a researched and planned process which is crucial for social
transformation, operating through three main strategies:
 Advocacy to raise resources and political and social leadership commitment for development
goals; Advocacy is a continuous and adaptive process of gathering, organizing and formulating
information into argument, to be communicated through various interpersonal and media
channels, with a view to raising resources or gaining political and social leadership and
commitment for a development program, thereby preparing a society for its acceptance.’
 Social mobilization for wider participation and ownership; and Social Mobilization is the process
of bringing together all feasible and practical inter-sectoral social partners and allies to determine
felt-need and raise awareness of, and demand for, a particular development objective. It involves
enlisting the participation of such actors, including institutions, groups, networks and
communities, in identifying, raising, and managing human and material resources, thereby
increasing and strengthening self-reliance and sustainability of achievements.’
 Program communication for changes in knowledge, attitudes and practices of specific participants
in programs; Program communication is a research-based, consultative process of addressing
knowledge, attitudes and practices through identifying, analyzing and segmenting audiences and
participants in programs and by providing them with relevant information and motivation through
well-defined strategies, using an appropriate mix of interpersonal, group and mass-media
channels, including participatory methods.’
When combined with strategies for the development of appropriate skills and capacities, and the provi-
sion of an enabling environment, communication plays a central role in positive behavior develop-
ment, behavior change and the empowerment of individuals and groups.”
31.6.Partnerships
You do not have to go it alone in a DRC campaign. Often, partnerships are developed which are very
fruitful because they --
 Reduce the cost of the project to you
 Develop long-term and more sustainable solutions if several partners are involved
 Local partners particularly assist with local contacts, culture, etc.
 They may take you out of the rut and routine that you are used to in doing your campaigns
153
Partnerships are not however without problems. They could not carry their load, their ideas and yours
might be so divergent that consensus becomes difficult, they may go off strategy and do their own
think, you may feel loss of ownership, etc. Ultimately you may feel you are spending more time on
managing partnership dynamics than if you had done it yourself. Hence great care needs to be
exercised in selecting partners. Research the partnership choices you have:
 Look for what mutual benefits can accrue from each
 Look at financial stability
 Look for compatibility of personalities
 Look for compatibility of ideas and approaches
 Look for complementary skills
Once you have selected a partner(s), very clearly define the roles and responsibilities of each in terms
of work, finances, etc. This is critical to forestalling problems.
31.7.DRC Budget
The budget for your DRC campaign is an important consideration. Sometimes you may have a defined
budget within which you have to develop your DRC campaign. Thinking through your campaign in a
systematic fashion through objectives may in fact assist you to exploit your predetermined budget to its
fullest. You may need to make some hard choices, but if they are based on good strategy they may be
easier to make.
On the other hand, you may be able to develop a DRC campaign and develop a budget for it. In this
case, the systematic planning of your DRC campaign will provide a strong rationale for each of your
decisions and therefore for your budget making it easier for you to convince potential donors or gov-
ernment agencies to support your project.
At the same time recognize that budgets are always limited. Responsible budgeting and ultimately re-
sponsible spending of budgets will increase your credibility.
Also recognize, that the process of budgeting is not necessarily linear. That is, you may not be able to
plan a full campaign and then begin the budgeting. Often as you are considering options, you will have
to get rough quotes to compare and make decisions. Sometimes this assists in developing solutions you
may not have thought of earlier. Local input is of very great importance at this stage in finding alterna-
tives, in exploring options, and in getting good deals because of local relationships.
As you budget, make sure you consider every expense. A DRC campaign plan is therefore very useful
in ensuring that you do not overlook any expenses. Remember the tangible items, such as X number of
booklets, may be easy to price. It is the intangibles such as time spent on preparing for an event that
may tend to be poorly budgeted. Take a step-by-step approach, creating a line item for each step. Later
after each step has been listed and costed chronologically, you may regroup into major categories that
belong together. The chronological method keeps you from leaving anything out. A media budget re-
cap sheet might also be very helpful for your media buys. This sheet should show spending for each
medium and total media spending.
154
As you budget, do not forget that you may find local donors who may be willing to provide some items
free either because it is a good cause or for the publicity that your campaign might be able to generate
for them. The media might be willing to provide free space and time for a good cause. Remember that
developing a relationship with a vendor might also assist in reducing costs. Further, bulk business with
a vendor is another possibility to explore. The advantage of this is that it gets the local community in-
volved in the effort.
The basic idea in budgeting is to achieve maximum benefit from the least amount of money. This is a
good principle to keep in mind but a difficult one to measure and operationalize. Another similar prin-
ciple is that there is a level beyond which spending more money will not help. There are some people
you are never going to reach with your message, there are others you are going to reach but with no ef-
fect on them. In both of these matters, past experience can play a critical role.
Some of the major expenses in a campaign are:

 Materials - design and production (includes writing, photographing, illustrations, layout, video,
audio, web sites, talent, etc.)
 Media Buys - The time and space you will buy on various channels
 Personnel - Project implementation once the campaign is planned; research implementation at
various stages starting with the situation analysis to post-campaign evaluation; training for
research, project implementation, etc.; travel expenses
 Contingency - Generally set aside a percent (usually 10%) for contingencies.
32.TECHNICAL ANNEX 21: TRAINING – EARLY WARNING

Goal (Session 6)
To provide an understanding of early warnings
Learning Outcome
 The participant will be able to formulate strategies for dissemination of early warnings to the
community
Learning Objectives

 Discriminate between hazard specific warnings
 Formulate and implement appropriate warning dissemination through the community
155
32.1.Classification of disasters for early warning activities
Disasters can be classified by nature, timing, predictability, response time and type of impact.
Table 39 Disasters according to timing

Slow Quick
Drought Flood
Famine Cyclone
Coastal Erosion Earthquake
Tsunami
Landslide
Table 40 Disasters according to predictability

Predictable Unpredictable
Drought Earthquake
Flood Earthquake
Coastal Erosion Landslide
Cyclone
Table 41 Disasters according to response time

Long response time 2-3 days response time 1-2 hours Response No response time
or less time or less
Drought Cyclone Tsunami Earthquake
Famine Floods Flash Flood
Landslide
Pre-impact/response
• Forecast
• Early warning
• Preparedness
• Tracking/monitoring approach of disaster
• Alertness/evacuation
32.2.Warning Systems
Forecasting helps to issue early warnings of disasters; the warning should be communicated to the
community through a warning system. This may be
 Alarms – like a fire alarm
 Sirens
 Public Announcements through Radio, TV (Cyclones, Floods, Landslides)
 Traditional communication systems in villages (temple bell, drums, hoisting of flags etc.)
156
32.3.Important Issues
 It must be noted that opportunity for warning does not exist in all cases.
 Indiscriminate warnings may result in non-responsiveness of people.
 A designated responsible person must always issue warnings
 All warning systems and technologies must be maintained in working condition and checked
regularly
 Communities in disaster prone areas must be made aware of the warning systems.
 Alternative warning systems must be kept in readiness in case of technological failure or power
failure.
 Only the delegated person will issue warning.
 All available warning systems should be used. Each warning system has limited reach and
multiple warning systems will help reinforcement.
 The warning should be very clear about the severity, the time frame and the area that may be
affected to the extent possible.
 Community should be made aware of the DO’s and DON’T’s after the warning as a
preparedness activity well ahead.
 Warnings should not evoke panic. It must be devoid of emotions.
 Rumor control action should be taken.
 Wherever possible get the assistance of community leaders and organized groups within the
community.
 Latest situation reports must follow the warning.
 If the disaster threat fades away an all-clear signal must be given.
 Evacuation
 It is important to understand the nature of the threat and the procedures to be adopted.
 All agencies involved in evacuation must have a common understanding of their roles and
responsibilities in order to avoid confusion and panic.
 Different situations demand different priorities and hence the responsibility of ordering an
evacuation is assigned to different agencies. The EOC, the Divisional Secretariat, Police and Fire
Brigade will only order evacuation.
 Evacuation should involve community leaders.
 All evacuations must be reported to the Divisional Secretariat.
157
32.4.Evacuation Preparedness Responsibility
Table 42 Evacuation preparedness responsibility

Preparedness/ Function Responsibility
 To warn people about the impend-  DS, GN. Police, Fire Brigade
ing danger & to leave for safer  Irrigation (flood)
places  Met Dept. (flood, cyclone)
 NBRO (landslides)
 Health (epidemics)
 Police (Road Accidents, Industrial
and Chemical Accidents)
 Geological Survey & Mines Bu-
reau, GSMB (E’quakes, Tsunami)
 Leading NGOs
 To co-ordinate with State-Civil De-  District/Divisional Secretaries
fense/Police/NGOs/CBOs  Police
 Leading NGOs/CBOs
 Arrangement of Boats/ vehicles etc  District/Divisional Secretaries
for speedy evacuation  Police/Army/Navy/Air Force
 Evacuate people of stranded areas  Leading NGOs/CBOs
 Organize trained task force mem-
bers
 Maintaining Law and Order  District/Divisional Secretaries
 Police/ Army/Navy/Air Force
 Divisional Sec., GN,
 Mobilize people to go to safe places  Village Disaster Management Com-
 Deployment of Boats for evacua- mittee
tion  Village Volunteers
 Police/Army/Navy/Air Force
158
33.TECHNICAL ANNEX 22: TRAINING – RESPONSE PLANNING

Goal (Session 7)
To provide ability for Response Planning
Learning Outcomes
 The participant will be able to formulate a Community Response Plan in collaboration with the
community
Learning Objectives

 Discriminate between House Hold Response Plan, Institutional (e.g. School) Response Plan and
Community Response Plan
 Understand the elements of an Evacuation Plan
 Formulate an Evacuation Plan
 Conduct an Evacuation Drill
 Evaluate the effectiveness of the Drill
33.1.Introduction
Response planning helps to prepare people to face a disaster situation. Each District with the leadership
of the District Secretary will form a District Disaster Management Committee (DDMC). This will act
as the Emergency Management and Response committee in the aftermath of a disaster. Activities
identified for emergency response are
 Early Warning
 Evacuation
 Identification of camps for displaced persons
 Search and rescue
 Post disaster assessment
 Emergency relief
 Logistics and supply
 Communication and information management
 Survivor response and coping
 Safety
 Security
 Emergency operations management
The Emergency Operations Centre (EOC) will be the focal point of operations control with the District
Secretary, Divisional Secretaries, District/Provincial Heads of Departments and Government Agencies
and other major stakeholders.
159
Response planning can be done at several levels.

 Family
 Institutional e.g. School, Office, Factory etc.
 Community and GN division
 Divisional
 District
 Provincial
 National
It is important that plans must conform to each other. Therefore it is advisable that all response plans
are vetted by the DDMC.
This session concentrates on the community level response. People can – and do – cope with disaster
by preparing in advance and working as a team. Helping people know what to do is the best protection.
33.2.What to Know? What to Plan
 Conduct a hazard assessment of the community to identify any hazards that may exist and
could cause an emergency. This is best done through a participatory approach. Involve commu-
nity leaders and the Grama Niladhari.
 Discuss the types of disasters that are most likely to happen. Understand what to do in each
case. Grama Niladhari can get help from the DDMC in the event you need outside help to do
this.
 Make a list of key personnel with contact information as well as contact information for local
NGO emergency responders, agencies and police.
 Identify vulnerable areas and mark households that should evacuate for safety
 Find out who needs help such as elderly, disabled persons, pregnant women, young children
and where they reside.
 Identify safe shelters for evacuation and best routes to take in the event of a disaster.
 Identify relief goods and services as well as logistics needed at evacuation shelters. Who would
be able to undertake them (resource mapping)?
 Identify persons who would manage evacuation shelters. What should they know and be able to
do?
 Learn about your community’s warning signals: what they sound like, how best to disseminate
them, and what people should do when they hear them.
 Ask about animal care after disaster. Animals may not be allowed inside
emergency shelters due to health regulations.
 Find out about the disaster plans at workplaces, children’s school or daycare centers.
160
33.3.Evacuation Policy and Procedure
All people involved in evacuation must have a common understanding of their roles and
responsibilities in order to avoid confusion and panicky behavior. The community plan must identify a
clear chain of command and designate a person authorized to order an evacuation. The EOC, District
Secretary or the Divisional Secretary, Police and Fire Brigade are the only authorized entities to order
an evacuation. Therefore the evacuation must be done in concurrence with any of them. The Police
must be notified immediately after the evacuation.
It is common practice to select a responsible individual to lead and coordinate your emergency plan
and evacuation. It is critical that community know who the coordinator is and understand that this
person has the authority to make decisions during emergencies. The coordinator should be responsible
for assessing the situation to oversee emergency procedures, notifying and coordinating with outside
emergency services.
 The community plan must address procedures for assisting people during evacuations, particularly
the sick, the elderly, the young and those with disabilities. Designate individuals as evacuation
wardens to help move people from danger to safe areas during an emergency. Generally, one war-
den for every 20 people should be adequate, and the appropriate number of wardens should be
available at all times.
 Encourage an entire family to evacuate together as a unit.
 Identify one or more assembly areas (as necessary for different types of emergencies and available
time) where people will gather first before proceeding to shelters and a method for accounting for
all persons worked out.
Accounting for all persons following an evacuation is critical. Confusion in the assembly areas
can lead to delays in helping disabled, or unnecessary and dangerous search-and-rescue
operations. To ensure the fastest, most accurate accounting of your community members, consider
taking a head count after the evacuation. The names and last known locations of anyone not
accounted for should be passed on to the official in charge.
 Amount of time needed for evacuation will depend on the disaster.
 As much as possible, the shelter sites should be within walking distance.
 Show evacuation routes with proper sign boards for different types of disasters. The extent of
evacuation may be different for different types of hazards.
 Evacuation routes should be well lit at night, wide enough to accommodate the number of evacu-
ating personnel, unobstructed and clear of debris at all times, and unlikely to expose evacuating
personnel to additional hazards.
33.4.House Hold Level Preparation
 Wear long-sleeved shirts, long pants and sturdy shoes so you can be protected as much as possi-
ble. Avoid dresses that impede speedy walking where necessary.
 Turn off electricity and water. Turn off electricity at the main fuse or breaker, and turn off water
at the main valve. Turn off gas cylinder.
 Take your disaster supplies kit.
If you have only moments before leaving, grab these things and go! Have them packaged in
161
anticipation of evacuation!
 Flashlight with plenty of extra batteries
 Battery-powered radio with extra batteries
 First aid kit
 Prescription medications in their original bottle, plus copies of the prescriptions
 Eyeglasses (with a copy of the prescription)
 Water (at least one gallon per person is recommended; more is better)
 Foods that do not require refrigeration or cooking
 Items that infants and elderly household members may require
 Medical equipment and devices, such as dentures, crutches, prostheses, etc.
 Change of clothes for each household member
 Sleeping bag or bedroll and pillow for each household member
 Checkbook, cash, and credit cards
 Lock your home.
 Set pets and livestock free. They will not be allowed inside shelter.
 Use identified evacuation routes — don't use shortcuts because certain areas may be impassable
or dangerous. Stay away from downed power lines.
If warning time is adequate
 Make arrangements for pets and livestock accommodation

 If high winds are expected, cover the outside of all windows of your home. Use shutters that are
rated to provide significant protection from windblown debris, or fit plywood coverings over all
windows.
 Protect your valuables
o Move objects that may get damaged by wind or water to safer areas of your home. Move
television sets, computers, stereo and electronic equipment, and easily moveable
appliances like a microwave oven to higher levels of your home and away from windows.
Wrap them in sheets, blankets, or polythene.
o Make a visual or written record of all of your household possessions. Record model and
serial numbers. This list could help you prove the value of what you owned if those
possessions are damaged or destroyed, and can assist you to claim deductions on taxes.
o Do this for all items in your home, including expensive items such as sofas, chairs, tables,
beds, chests, wall units, and any other furniture too heavy to move.
o Store a copy of the record somewhere away from home, such as in a safe deposit box.
33.5.Training and Drills
 Does the community plan identify how and when community members will be trained so that
they understand the types of emergencies that may occur, their responsibilities and actions as out-
lined in the plan?
 Does the plan address if and how often drills will be conducted? Once you have reviewed your
emergency action plan and everyone has had the proper training, it is a good idea to hold practice
drills as often as necessary to keep people prepared. Include outside resources such as
162
- fire and police departments when possible. After each drill, evaluate the
- effectiveness of the drill. Identify the strengths and weaknesses of your plan and work to
improve it.
34.TECHNICAL ANNEX 23: TRAINING –MONITORING AND EVALUATION
Goal (Session 8)
To create an understanding of the need for and methods of monitoring and evaluation.
Learning Outcomes
 The participant will be able to understand the importance of monitoring and evaluation and of
techniques for conducting them and using their results.
Learning Objectives
 The participant will learn to

 Explain the difference between monitoring and evaluation
 Understand the importance of these two aspects to measure success of a project
 Apply monitoring techniques (How? Who? What?)
 Apply evaluation techniques (How? Who? Scope? When?)
 Explain monitoring and evaluation terminology, techniques, and designs
 Understand the need for and techniques for documentation.
34.1.Definition of Monitoring and Evaluation
Monitoring
Monitoring refers to tracking the progress of your campaign as it is being implemented. It is concerned
more with implementation details, problems and adjustments in them, and not essentially with whether
the campaign’s communication and action objectives are being met.
Evaluation
Whether program objectives are being met or not falls under the purview of evaluation. Evaluation is
assessing gains in the effects desired. Evaluation is therefore conducted against the communication and
action objectives.
Difference between Monitoring and Evaluation
While some people do not make this distinction between monitoring and evaluation and often include
evaluation in monitoring, it is a useful difference to keep in mind. Monitoring is essentially done to
make sure everything (activities, visits, distribution, media placement, etc.) is taking placed as it was
163
planned. Evaluation, on the other hand, is to assess whether these things (activities, visits, distribution,
media placement, etc.) resulted in the effects desired, i.e., whether knowledge increased, for example.
34.2.Place of Monitoring and Evaluation in DRC
Monitoring
Monitoring will enable you to keep ahead of the campaign, ensure quality control, keep your
staff/volunteers/implementers alert and on the go, and such. When a campaign is very tightly planned
so that certain activities logically follow others, then failure in implementing one aspect of the
campaign affects all subsequent stages.
Evaluation
While monitoring is met with less apprehension and is generally included in DRC campaigns,
evaluation is often not conducted at all or at least not conducted as systematically as necessary. This
may be due to the fear that results may not be as positive as expected or due to the lack of realization
of the importance of evaluation or simply due to lack of budget or even skills. None of these are really
good reasons for not conducting an evaluation. Further, in this day and time, funding agencies are
more and more focused on accountability, and evaluation ensures accountability.
There are several good reasons to do an evaluation:

 Primarily, results are useful in your own campaign. Respond to them in real time. If you find that
KASP levels are not rising at the rate expected, find out the source of the problem (lack of
distribution of message, garbled message, inaccessibility to message, etc.) and fix it. The problem
may lie in the message, in the people or media delivering the message, in the timing, and in a host
of other places. Identify these causes and move fast to eliminate them so that the campaign is back
on track as soon as possible and problems do not snowball and have a chain reaction. In essence,
making these adjustments as problems are identified is iterative.
 Results will be useful to others working in the same field. Particularly in the field of disasters,
literature on DRC campaigns is scarce.
 Unexpected things can happen and derail even the best planned campaigns, but if your campaign
was informed by research at each stage, then the results of the evaluation should most likely
indicate strides in achieving objectives. This will boost your confidence and convince others of the
merits of your work.
A certain amount of planning and foresight will keep evaluation from becoming an object of fear or a
burden.
 Set realistic communication and action objectives at the start of the campaign. Changes in most
variables do not occur easily and overnight and are only incremental in nature not drastic. Most
people make the mistake of setting unreal goals or confusing expected results from
communication and action objectives. Remember communication objectives can make people
aware or create knowledge, recall, etc., but may not result in action unless other elements are in
place in the campaign.
 Budget for evaluation from the very start. Evaluation is important enough that you must plan for it
in terms of budget. Remember this involves not just money but also staff time and expertise.
164
 Make the budget large enough to do a systematic evaluation. Do not conduct a half-baked
evaluation. You will then end up making unfounded conclusions about the success or failure of
the campaign.
 Develop skills in evaluation or engage an expert to conduct it. Evaluation is serious business. It
requires skill if it is to be done right. Either develop your own skills or engage an expert to
conduct it.
If your campaign is not having an effect, an evaluation will indicate that and you may be able to make
improvements in time to save it. Otherwise you will continue to spend resources on an ineffective
campaign.
34.3.Monitoring Plan and Mechanism
While it may not be possible to anticipate everything, it is a good idea to put a monitoring plan in place
before campaign start, and to begin monitoring activity from the beginning of the campaign. This plan
should detail who will check on what activity (all major activities must be tracked), when (at very
regular intervals such as daily/weekly if the activity is ongoing or at time of activity if activity is a one-
timer), where (actual site visits, from office, etc.), and how (see below). A chart that outlines this will
be very useful in keeping staff and monitoring on track.
The monitoring mechanisms (the “how” of the monitoring plan) might be simply counting brochures,
or making sure the posters were distributed on a certain date, or ensuring that the media ran the
campaign notice of a contest in time. Here are some mechanisms that may be used:
 Materials inventory
 Materials placement audit
 Materials distribution audit
 Media logs, tear sheets
 Budget assessment
 Timeline assessment
 Staff meetings and reports
A very useful tool is a calendar with considerable detail entered (when, where, how many, etc.) on
every activity that needs to take place. Such a calendar will be useful to implementers as well as to the
monitoring staff. A master calendar in the head office with copies for implementers (volunteers,
officials, etc.) would ensure shared knowledge of the timing, type and scope of each activity. Each one
of your partners, implementers, volunteers, etc., would know how crucial they are to the complete plan
and how the success of other activities is dependent on their success in delivering on their
responsibilities. The calendar also makes very clear the consequences of failing to get any one activity
done in time or in the scope planned. Similarly, monitoring staff can identify the activities to be
monitored on this calendar and use it as a scheduling guide for their monitoring activities. Calendar
apart, staff will have to be trained to conduct the monitoring and to file monitoring reports.
34.4.Evaluation Terminology
There is considerable difference in the use of terminology for evaluation. Simply, however, evaluation
may take place:
165
 During the campaign (concurrent evaluation) and at its finish (post-campaign evaluation).
 Concurrent evaluation may be one time (mid-campaign, for example) or at multiple points in time
(critical moments).
 Post-campaign evaluation may be done one time (immediately on close of campaign) or at
repeated intervals after post-campaign evaluation. The latter is done to see if campaign effects are
long-term but it is also rather expensive to continue evaluation.
 Terminology you may be familiar with is:
 Process evaluation (concurrent evaluation) and summative evaluation (post-campaign evaluation).
The term process evaluation is sometimes also used to refer to monitoring activities.
 Summative evaluation is of two types: outcome and impact.
 Outcome evaluation is the linking of program objectives and gains made in desired effects. It
assesses individual level changes. Conducting individual level data as part of outcome evaluation
has many advantages other than providing a direct measure of campaign success against
objectives. Such data can be analyzed in many creative ways to see which groups had more or less
gains. The groups may be defined by gender, education, psychographic and many other
characteristics. This will provide invaluable data for other phases of the campaign, for writing
proposals to funding agencies, and for others in the field.
 Impact evaluation focuses on long-term impact. It asks questions such as: Was there a reduction in
number of people killed?
34.5.Evaluation Plan
Strategic Planning of Outcome Evaluation - Much evaluation that is done may be anecdotal in nature.
This is not evaluation in the real sense of the term. Bias may enter in the selection of people who tell
anecdotes or in the reporting of anecdotes. Similarly, post-hoc counting of brochures distributed or
noting the enthusiasm of people at an activity is not evaluation. Rather rigorous, multi-method (called
triangulation) evaluation must be conducted against the communication and action objectives defined
earlier in the campaign. This is strategic research as opposed to ad hoc, haphazard evaluation research.
Formulate your evaluation plan before the start of the campaign. The plan for your evaluation will
depend on the length of your campaign and the resources you have.
The plan should detail when (how often, exact time period), where (which sample), and how (which
methods) you will conduct the evaluation(s).
For example, if your campaign is one year long and you have necessary resources, you may plan to
conduct two concurrent evaluations, one every four months and a post-campaign evaluation. You may
randomly pick one or more of several target communities or you may select randomly respondents
from each target community as sample. You may decide to conduct a survey and focus groups at each
time. You may decide on a panel survey (same sample of people over the three evaluations) or a trend
(different sample each time) and on conducting at least four focus groups in each sample community.
Critical are the communication and action objectives set earlier. These define what you set out to
achieve and therefore measurement of the success of your DRC campaign should be made to
demonstrate if these objectives were in fact achieved.
To be able to measure results against the objectives set early in the campaign, before any outcome
evaluation is done, baseline data must be collected. To do this, develop a questionnaire based largely
on the communication and action objectives of the campaign. That is, the questionnaire must capture
166
the current (pre-campaign) levels of the effects desired (e.g., knowledge, attitude, Skills and practice
levels if these were the targeted effects). Considerable care must be used in developing this
questionnaire because it will be used for the concurrent and post-campaign evaluations; questions may
be added but the ones already in the questionnaire should not be changed.
34.6.Outcome Evaluation Designs
Generally survey and other qualitative methods such as focus groups, in-depth interviews, target
audience diaries, and observation are used in outcome evaluation. Experiments, particularly laboratory
experiments, are hard to conduct because of the field nature of DRC campaigns which makes control
of rival explanations difficult.
While experiments may be difficult in DRC, field experiments are possible if you can identify a
control community that is very similar to your intervention community and has not had any spillover
effects from the campaign. That is, the control community must have the same characteristics that your
target group has in terms of demographics, psychographics, geographies, KASP levels, etc. (in essence
any variable critical to the campaign) and also not have been indirectly exposed to any campaign
messages. Here is one possible design:
Intervention Community Pre-campaign measure CM 1 CM 2 PCM 1

PCM2
Control Community PCM
Note that if resources require that you make changes in the design of your survey or field experiment,
as far as possible do not remove the pre-campaign measure.
34.7.Community Indicators
Such outcome evaluation though can be expensive. Sometimes when resources are very limited,
community level indicators may be used (this is not a recommended measure however other than as
supplemental to the individual level data). Community level indicators are generally garnered through
observation. You may observe if people are learning how to swim or are strengthening homes, how
many people are contacting offices for information, whether the media are covering the campaign and
the topic of disaster in particular, whether legislative or policy changes are in the making, etc. Even
with this kind of data, it is important not to bring in researcher bias in interpretation.
34.8.Steps in Evaluation Research
 Specifically, the steps in your evaluation plan will be:

 Determining the data to be collected based on your communication and action objectives
 Selecting a method(s) to use for the evaluation.
 Designing your data collection instrument
 Collecting and analyzing your data
 Writing a final report
167
Follow this outline in writing your research report but remember to keep the report user friendly (see
attached):
 Introduction: provide the general context within which the DRC campaign has been initiated,
elements of the DRC campaign, and its specific communication and action objectives.
 Method: provide details on the method(s) picked (e.g., survey), why; the design picked (e.g.,
longitudinal), why; the data collection instrument picked (e.g., questionnaire), why; a description
of the variables and how they are measured, why; and population and sample, why. Note that
these details should be provided both for the outcome and impact evaluations as well as for
concurrent and post-campaign evaluations.
 Findings: provide a description of the sample particularly in demographic terms; provide results in
terms of KASP and other targeted effects (including attitude to messages, channels used for
delivery, etc.). Here you may outline both outcome and impact evaluation results.
 Conclusions: provide a summary of what worked and what did not based on the findings.
 Recommendations: provide recommendations based on these results for others and your own
future work.
34.9.Documentation
Monitoring and evaluation are both related to documentation. Documentation is the recording of the
experience of project implementation as well as of results of the evaluation. Reasons to document are:
 Provide secondary information to others in the field saving them time and effort
 Provide a proof of performance of your efforts
 Provide materials for display, presentation, academic papers, etc.
 Provide funding agencies with an overview of your capabilities
 Use as a tool to draw in partners, volunteers, etc.
 Use as a too to boost pride among your staff and others involved in the project
 Use as a tool to send to the media
Staff needs to be socialized in the importance of documentation and trained to contribute to this
documentation effort. Together, make a documentation plan so that the effort is not haphazard. Decide
what will be recorded in text, pictures, audio and video, at what times within the progress of the
campaign and its completion, and who will be responsible for this. This responsibility includes writing
reports, clipping newspaper stories and getting taped copies of other press coverage, storing a few
copies of all campaign materials, plans, etc., and such.
168
35.APPENDIX A: MAP OF SRI LANKA WITH 32 COASTAL COMMUNITIES

The Sarvodaya villages marked on the map were selected such that 16 are Sarvodaya 1, 2, & 3 stage
communities and the remaining 16 are stage 4 & 5. Table 5 in Technical Annex 4 complies with the set
of villages identified on this map.
Figure 45 Locations of 32 HazInfo project participating communities.
169
36.APPENDIX B: DATA OF RESEARCH MATRIX ALLOCATION
Table 43 Allocation of ICTs to Communities with training and organizational capacity

ERP Develop
District Trained/ ment Committee ICT
NO Name Community Name Untrained Stage Name Name
1 Abeysainghepura Trained 2 None None
Ampara Sarvodaya shramadana so-
2 Panama North Trained 4 AREA
ciety
3 Modara Untrained 2 None None
Colombo "SAMUDURA" Sarvodaya
4 Moratuwella Trained 2 AREA
S.S
5 Sarvodaya shramadana so-
Brahammanawatta Untrained 4 AREA
ciety
6
Urawatta Trained 2 Sarvodaya Center VSAT
7 Sarvodaya shramadana so- AREA
Galle Velhengoda Trained 4
ciety MOP
Venamulla Untrained 4
ciety RAD
Indivinna Untrained 4 FXP
ciety
Modarapelessa Trained 2
11 ciety RAD
Hambantota
Modaragama Trained 4 Sarvodaya D/C VSAT
Samodhagama Untrained 4 MOP
ciety
13 AREA
Munai Untrained 2 -
MOP
14 Jaffna Sarvodaya shramadana so- AREA
Thondamanar Untrained 2
ciety RAD
Kalmunai II Trained 4
ciety RAD
Ninthavur Trained 2 MOP
17 ciety
Kalmunai Oluvilu Untrained 2 FXP
ciety
Sarvodaya shramadana so- AREA
Karathivu Untrained 2
19 ciety FXP
Thambiluvil Untrained 2 Sarvodaya Thambiluvil MOP
Diyalagoda Trained 4 MOP
ciety
Kalutara Maggona Untrained 2 AREA
ciety
22
Talpitya Untrained 4 None NONE
170

Kottegoda Untrained 4
ciety FXP
24
Madiya Untrained 2 Sarvodaya D/C MOP
25
Mirissa South Trained 4 None NONE
26 Matara
Sarvodaya shramadana so- AREA
Talalla South Untrained 4
ciety MOP
Wattegama South Trained 2
ciety FXP
28 AREA
Samudragama Trained 4 Sarvodaya Shanthi Sena
FXP
29 Trincomalee
Thirukkadalar Trained 2 Sarvodaya Trincomalee FXP
30 Maangadu Untrained 4 Sarvodaya Shanthi Sena MOP
31 Periyakallar Trained 4 Sarvodaya Shanthi Sena FXP
Batticaloa AREA
32 Palamunai Trained 2 Sarvodaya Shanthi Sena
MOP
37.APPENDIX C: DATA OF HIH OPERTIONAL
Table 44 live-exercise data for function: Download, Acknowledgment, & EOI completion
0 Exercise SW Download Acknowledge EOI
NO Date Name Start End Eff Start End Eff Start End Eff
11-
1 0.00% 0.00% 10:10 10:14 0.95
Dec-06
11-
2 ANNY 0.00% 0.00% 0.00%
Dec-06
4- 1.0
3 ANNY 10:30 10:31 0.98 10:34 10:34 10:34 10:36 0.97
Mar-07
4- 1.0
4 ANNY 10:30 10:31 0.98 10:34 10:34 10:34 10:36 0.97
Mar-07
26-
5 ANNY 0.00% 0.00% 0.00%
Nov-06
26-
6 IPAS 0.00% 0.00% 0.00%
Nov-06
26-
7 ANNY 0.00% 0.00% 0.00%
Nov-06
26-
0 DEWN 0.00% 0.00% 0.00%
Nov-06
26- ANNY
8 0.00% 0.00% 0.00%
Nov-06 DEWN
9 0-Jan-00 CALL 0.00% 0.00% 0.00%
9-
10 ANNY 10:16 10:17 0.98 10:17 0.00% 10:17 10:20 0.96
May-07
11 9- DEWN 10:16 10:17 0.98 10:17 0.00% 10:17 10:20 0.96
171
May-07
9-
12 DEWN 10:16 10:17 0.98 10:17 0.00% 10:17 10:20 0.96
May-07
ANNY
13 0.00% 0.00% 0.00%
DEWN
ANNY
14 0-Jan-00 0.00% 0.00% 0.00%
DEWN
17- ANNY 1.0
15 9:06 9:06 1.0 9:06 9:06 9:06 9:08 0.97
Mar-07 DEWN
17- 1.0 1.0
16 DEWN 9:06 9:06 9:06 9:06 9:06 9:08 0.97
Mar-07
17- 1.0 1.0
17 CALL 9:06 9:06 9:06 9:06 9:06 9:08 0.97
Mar-07
17- ANNY 1.0 1.0
18 9:06 9:06 9:06 9:06 9:06 9:08 0.97
Mar-07 CALL
17- 1.0 1.0
19 DEWN 9:06 9:06 9:06 9:06 9:06 9:08 0.97
Mar-07
11- 1.0
20 DEWN 10:22 10:22 0.00% 10:23 0.00%
Mar-07
11- 1.0
21 ANNY 10:22 10:22 0.00% 10:23 0.00%
Mar-07
22 9-Apr-06 10:22 10:22 1.0 0.00% 10:23 0.00%
11- ANNY
23 0.00% 0.00% 10:10 10:14 0.95
Dec-06 CALL
25-
24 DEWN 0.00% 0.00% 10:10 10:14 0.95
Feb-07
11-
25 0.00% 0.00% 10:10 10:14 0.95
Dec-06
11- ANNY
26 0.00% 0.00% 10:10 10:14 0.95
Dec-06 DEWN
11- ANNY
27 0.00% 0.00% 10:10 10:14 0.95
Dec-06 CALL
10- ANNY 1.0
28 10:08 10:08 0.00% 10:10 10:11 0.98
Mar-07 CALL
10- 1.0
29 CALL 10:08 10:08 0.00% 10:10 10:11 0.98
Mar-07
25-
30 DEWN 0.00% 0.00% 0.00%
Feb-07
25-
31 CALL 0.00% 0.00% 0.00%
Feb-07
25-
32 ANNY 0.00% 0.00% 0.00%
Feb-07
172
Table 45 Live-exercise data for function: Approval Issue and overall efficiency
0 Exercise SW Approval Issue Exercise Overall
NO Date Name Start End Eff Start End Eff Duration Eff
1 11-Dec-06 0.00% 0.00% 1:30 0.00%
2 11-Dec-06 ANNY 0.00% 10:36 10:42 93.33% 1:30 93.33%
3 4-Mar-07 10:36 10:39 96.67% 0.00% 1:30 0.00%
4 4-Mar-07 ANNY 10:36 10:39 96.67% 10:39 10:45 93.33% 1:30 93.33%
26-
5 ANNY 0.00% 0.00% 1:30 0.00%
Nov-06
26-
6 IPAS 0.00% 0.00% 1:30 0.00%
Nov-06
7 26- ANNY 0.00% 0.00% 1:30 0.00%
0 Nov-06 DEWN 0.00% 0.00% 1:30 0.00%
26-
8 ANNY 0.00% 0.00% 1:30 0.00%
Nov-06
26- DEWN 0.00% 0.00% 1:30 0.00%
09 Nov-0600 CALL 0.00% 0.00% 1:15 0.00%
ANNY 10:20 10:21 98.67% 10:28 10:50 70.67% 1:15 70.67%
10 9-May-07 0.00% 0.00% 1:15 0.00%
11 9-May-07 DEWN 10:20 10:21 98.67% 10:23 10:25 97.33% 1:15 97.33%
12 9-May-07 DEWN 10:20 10:21 98.67% 10:25 10:28 96.00% 1:15 96.00%
ANNY 0.00% 0.00% 0:00 0.00%
13 00 DEWN 0.00% 0.00% 0:00 0.00%
ANNY 0.00% 0.00% 0:00 0.00%
140 000 DEWN 0.00% 0.00% 0:00 0.00%
ANNY 9:08 9:10 97.78% 9:10 9:18 91.11% 1:30 91.11%
15 17-Mar-07 DEWN 0.00% 9:56 9:58 97.78% 1:30 97.78%
16 17-Mar-07 DEWN 9:08 9:10 97.78% 0.00% 1:30 0.00%
17 17-Mar-07 CALL 9:08 9:10 97.78% 0.00% 1:30 0.00%
ANNY 9:08 9:10 97.78% 9:10 9:18 91.11% 1:30 91.11%
18 17-Mar-07 CALL 0.00% 0.00% 1:30 0.00%
19 17-Mar-07 DEWN 9:08 9:10 97.78% 0.00% 1:30 0.00%
20 11-Mar-07 DEWN 10:24 10:25 98.89% 10:28 10:32 95.56% 1:30 95.56%
21 11-Mar-07 ANNY 10:24 10:25 98.89% 10:25 10:28 96.67% 1:30 96.67%
22 9-Apr-06 10:24 10:25 98.89% 0.00% 1:30 0.00%
ANNY 0.00% 10:36 10:42 93.33% 1:30 93.33%
23 11-Dec-06 CALL 0.00% 10:55 11:02 92.22% 1:30 92.22%
24 25-Feb-07 DEWN 0.00% 10:51 10:53 97.78% 1:30 97.78%
25 11-Dec-06 0.00% 0.00% 1:30 0.00%
ANNY 0.00% 10:36 10:42 93.33% 1:30 93.33%
26 11-Dec-06
DEWN 0.00% 10:51 10:53 97.78% 1:30 97.78%
173
ANNY 0.00% 10:36 10:42 93.33% 1:30 93.33%

27 11-Dec-06 CALL 0.00% 10:55 11:02 92.22% 1:30 92.22%
ANNY 10:11 10:15 95.56% 10:15 10:19 95.56% 1:30 95.56%
28 10-Mar-07 CALL 10:11 10:15 95.56% 10:20 10:22 97.78% 1:30 97.78%
29 10-Mar-07 CALL 10:11 10:15 95.56% 10:33 10:34 98.89% 1:30 98.89%
30 25-Feb-07 DEWN 0.00% 0.00% 1:30 0.00%
31 25-Feb-07 CALL 0.00% 0.00% 1:30 0.00%
ANNY 0.00% 0.00% 1:30 0.00%
32 25-Feb-07 DEWN 0.00% 0.00% 1:30 0.00%
Approval Issue
99.1 69.91
38.APPENDIX D: DATA OF EFFICIENCY MEASURE FROM LIVE-EXERCISES
Table 46 Efficiency data for HIH Monitor and ICT Terminal Devices Pre Community
ICT Exercise HIH Monitor ICT Terminal Device
NO Sent Received Ack Eff Sent Received Ack Eff
Name Date
1 None 12/11/2006 10:19 10:21 0.00% 0.00%
2 AREA 12/11/2006 10:19 10:21 96.67% 10:24 10:24 98.89%
3 None 3/4/2007 10:10 10:35 0.00% 0.00%
4 AREA 3/4/2007 10:10 10:35 83.33% 10:49 10:50 87.78%
5 AREA 11/26/2006 10:25 10:35 87.78% 10:46 10:46 82.22%
6 VSAT 11/26/2006 10:25 10:35 0.00% 0.00%
AREA 10:25 10:35 61.11% 11:10 11:10 100.00%
7 MOP 11/26/2006 10:25 10:35 83.33% 10:50 10:50 100.00%
AREA 10:25 10:35 87.78% 10:46 10:46 97.78%
8 RAD 11/26/2006 10:25 10:35 92.22% 10:42 10:42 100.00%
9 FXP 1/0/1900 10:25 10:35 0.00% 0.00%
AREA 10:15 10:20 54.67% 10:54 10:54 85.33%
10 RAD 5/9/2007 10:15 10:20 0.00% 0.00%
11 VSAT 5/9/2007 10:15 10:20 0.00% 0.00%
12 MOP 5/9/2007 10:15 10:20 0.00% 0.00%
AREA 1/0/1900 0.00% 0.00%
13 MOP 0.00% 0.00%
1/0/1900
14 AREA 0.00% 0.00%
RAD 1900 0.00% 0.00%
0
AREA 9:15 9:16 97.78% 9:18 9:18 97.78%
15 RAD 3/17/2007 9:15 9:16 92.22% 9:20 9:23 96.67%
16 MOP 3/17/2007 9:15 9:16 88.89% 9:24 9:26 97.78%
17 FXP 3/17/2007 9:15 9:16 88.89% 9:26 9:26 100.00%
174
AREA 9:15 9:16 97.78% 9:18 9:18 23.33%

18 FXP 3/17/2007 9:15 9:16 0.00% 0.00%
19 MOP 3/17/2007 9:15 9:16 0.00% 0.00%
20 MOP 3/11/2007 10:15 10:30 77.78% 10:50 10:50 97.78%
21 AREA 3/11/2007 10:15 10:30 97.78% 10:32 10:32 85.56%
22 None 4/9/2006 10:15 10:30 0.00% 0.00%
AREA 10:19 10:21 96.67% 10:24 10:24 98.89%
23 FXP 12/11/2006 10:19 10:21 84.44% 10:35 10:35 98.89%
24 MOP 2/25/2007 10:19 10:21 87.78% 10:32 10:32 94.44%
25 None 12/11/2006 10:19 10:21 0.00% 0.00%
AREA 10:19 10:21 96.67% 10:24 10:24 98.89%
26 MOP 12/11/2006 10:19 10:21 87.78% 10:32 10:32 95.56%
AREA 10:19 10:21 96.67% 10:24 10:24 98.89%
27 FXP 12/11/2006 10:19 10:21 78.89% 10:40 10:40 98.89%
AREA 10:20 10:23 98.89% 10:24 10:24 91.11%
28 FXP 3/10/2007 10:20 10:23 90.00% 10:32 10:32 100.00%
29 FXP 3/10/2007 10:20 10:23 90.00% 10:32 10:32 98.89%
30 MOP 2/25/2007 10:15 10:18 0.00% 0.00%
31 FXP 2/25/2007 10:15 10:18 0.00% 0.00%
AREA 10:15 10:18 97.78% 10:20 10:20 100.00%
32 MOP 2/25/2007 10:15 10:18 0.00% 0.00%
Table 47 Efficiency data for ICT Guardian and ERP Coordinator in Community
ICT Exercise ICT Guardian ERP Coordinator
NO Sent Eff Ack Eff Sent Eff Ack Eff
Name Date
1 None 12/11/2006 0.00% 10:36 10:43 47.78%
2 AREA 12/11/2006 10:24 10:25 86.67% 10:36 10:43 47.78%
3 None 3/4/2007 0.00% 10:47 10:51 63.33%
4 AREA 3/4/2007 10:50 11:00 94.44% 10:55 10:55 44.44%
5 AREA 11/26/2006 10:46 11:02 82.22% 11:02 11:02 100.00%
6 VSAT 11/26/2006 0.00% 0.00%
AREA 11:10 11:10 97.78% 11:12 11:13 74.44%
7 MOP 11/26/2006 10:50 0.00% 11:12 11:13 74.44%
AREA 10:46 10:48 95.56% 10:50 10:55 83.33%
8 RAD 11/26/2006 10:42 10:42 91.11% 10:50 10:55 83.33%
9 FXP 1/0/1900 0.00% 0.00%
AREA 10:55 11:05 80.00% 11:10 11:15 66.67%
10 RAD 5/9/2007 0.00% 11:10 11:15 66.67%
11 VSAT 5/9/2007 0.00% 10:40 10:41 66.67%
12 MOP 5/9/2007 0.00% 10:40 10:41 66.67%
AREA 0.00% 0.00%
13 MOP 0.00% 0.00%
175
AREA 0.00% 0.00%

14 RAD 0 0.00% 0.00%
AREA 3/17/2007 9:19 9:20 91.11% 9:27 9:30 68.89%
15 RAD 3/17/2007 0.00% 9:27 9:30 68.89%
16 MOP 3/17/2007 0.00% 10:26 10:29 67.78%
17 FXP 3/17/2007 0.00% 0.00%
AREA 10:26 10:27 96.67% 10:29 10:30 71.11%
18 FXP 3/17/2007 10:26 10:27 96.67% 10:29 10:30 71.11%
19 MOP 3/17/2007 10:27 10:29 95.56% 10:31 10:32 51.11%
20 MOP 3/11/2007 10:50 10:52 93.33% 10:56 10:58 73.33%
21 AREA 3/11/2007 10:32 10:45 83.33% 10:47 10:55 63.33%
22 None 4/9/2006 0.00% 0.00%
AREA 10:24 10:25 86.67% 10:36 10:40 62.22%
23 FXP 12/11/2006 10:35 10:36 98.89% 10:36 10:40 62.22%
24 MOP 2/25/2007 10:32 10:37 93.33% 10:38 10:47 58.89%
25 None 12/11/2006 0.00% 10:50 10:52 68.89%
AREA 10:24 10:25 84.44% 10:38 10:45 53.33%
26 MOP 12/11/2006 10:35 10:36 96.67% 10:38 10:45 53.33%
AREA 10:24 10:25 96.67% 10:27 10:30 44.44%
27 FXP 12/11/2006 10:40 10:41 87.78% 10:51 10:53 84.44%
AREA 10:27 10:32 88.89% 10:37 10:39 88.89%
28 FXP 3/10/2007 10:32 10:32 94.44% 10:37 10:39 88.89%
29 FXP 3/10/2007 10:32 10:33 94.44% 10:37 10:39 88.89%
30 MOP 2/25/2007 0.00% 0.00%
31 FXP 2/25/2007 0.00% 0.00%
AREA 10:20 0.00% 0.00%
320 MOP 2/25/2007 0.00% 0.00%
Table 48 - Efficiency data summary for all functions in Tables 20 & 21

ICT Exercise Exercise Hazard Exercise
End
NO Name Date Duration Detect Impact MTTF Comments
Time
ICTG Incompetent
in documenting
1 None 12/11/2006 11:23 1:23 10:00 11:30 92% mag, Delay in
ICTG contacting
ERPC
ICTG incompetent
in documenting
2 AREA 12/11/2006 11:23 1:23 10:00 11:30 92% msg, Delay in
ICTG contacting
ERPC
3 None 3/4/2007 11:20 1:20 10:00 11:30 89% Delay in HIH
176
Monitor retrieving
email for details
Delay in communi-
4 AREA 3/4/2007 11:45 1:45 10:00 11:30 100% ty receiving alert
from Modara
AREA had 2 bars
and didn't receive
alert because no
5 AREA 11/26/2006 11:02 0:47 10:15 11:45 52%
FEC, signal
strength must be >
4
6 VSAT 11/26/2006 11:45 1:30 10:15 11:45 100%
AREA Did not receive 1st
11:35 1:20 10:15 11:45 89%
alert, received 2nd,
7 11/26/2006 MOP Received
MOP
11:35 1:20 10:15 11:45 89% alert but siren not
heard by ICTG
AREA 2nd received
11:05 0:50 10:15 11:45 56%
AREA alert
8 RAD 11/26/2006 1st received RAD
11:05 0:50 10:15 11:45 56%
alert
9 FXP 1/0/1900 11:30 1:15 10:15 11:30 100%
AREA 11:35 1:20 10:15 11:30 100%
RAD was sent for
10 5/9/2007 repairs and had not
RAD 11:35 1:20 10:15 11:30 100%
received before
simulations
11 VSAT 5/9/2007 11:05 0:50 10:15 11:30 67%
RAD was used in-
stead of MOP,
12 MOP 5/9/2007 11:05 0:50 10:15 11:30 67%
could not get ap-
plet to work
AREA 0:00 0%
13 MOP 0:00 0%
AREA 0:00 0%
14 RAD /1900 0:00 0%
Delay in ICT-G re-
AREA 3/17/2007 9:55 0:55 9:00 10:30 61%
ceiving message
Did not receive
RAD & No GSM
15 Coverage bc of
RAD 3/17/2007 9:55 0:55 9:00 10:30 61% North-east Con-
flict, District Coor-
dinator Called with
land line
177
No GSM Coverage
bc of North East
16 MOP 3/17/2007 10:55 1:55 9:00 10:30 100% Conflict, District
Coordinator called
via land-line
No GSM Coverage
bc of North East
17 FXP 3/17/2007 10:55 1:55 9:00 10:30 100% Conflict, District
Coordinator called
via land-line
AREA Received AREA
alert after reassem-
bling the AREA,
No GSM Coverage
10:55 1:55 9:00 10:30 100%
bc of North East
Conflict, District
18 3/17/2007 Coordinator called
FXP via land-line
No GSM Coverage
bc of North East
10:55 1:55 9:00 10:30 100% Conflict, District
Coordinator called
via land-line
No GSM Coverage
bc of North East
19 MOP 3/17/2007 11:15 2:15 9:00 10:30 100% Conflict, District
Coordinator called
via land-line
Church bell and
Mosque speakers
20 MOP 3/11/2007 11:20 1:20 10:00 11:30 89% announcement not
heard by all in
community
21 AREA 3/11/2007 11:20 1:20 10:00 11:30 89%
Thalpitiya was ex-
cluded bc live-ex-
ercises were run
during ToT pro-
22 None 4/9/2006 11:30 1:30 10:00 11:30 100%
gram as well as by
Gov DMC and did
not want to exhaust
them
AREA 11:10 1:10 10:00 11:30 78%
23 FXP 12/11/2006 11:10 1:10 10:00 11:30 78%
24 MOP 2/25/2007 11:15 1:15 10:00 11:30 83%
178
25 None 12/11/2006 11:18 1:18 10:00 11:30 87%

AREA 11:20 1:20 10:00 11:30 89%
26 MOP 12/11/2006 11:20 1:20 10:00 11:30 89%
AREA 11:17 1:17 10:00 11:30 86%
27 FXP 12/11/2006 11:05 1:05 10:00 11:30 72%
AREA 10:47 0:32 10:15 11:45 36%
28 FXP 3/10/2007 10:47 0:32 10:15 11:45 36%
29 FXP 3/10/2007 10:47 0:32 10:15 11:45 36%
Communities un-
prepared to con-
duct live exercises;
30 MOP 2/25/2007 11:30 1:30 10:00 11:30 100%
North-East conflict
was a higher priori-
ty
Communities un-
prepared to con-
duct live exercises;
31 FXP 2/25/2007 11:30 1:30 10:00 11:30 100%
North-East conflict
ty
AREA Communities un-
prepared to con-
32 duct live exercises;
11:30 1:30 10:00 11:30 100%
2/25/2007 North-East conflict
0 MOP
ty
11:30 1:30 10:00 11:30 100%
39.APPENDIX E: DATA FOR CERTAINTY MEAURE OF ICT

Table 49 Signal strength measured in terms of enumerating bars for each deployment
0 ICT
No Deployment Min Max Mean Score
1 None - - - -
2 AREA 2.00 7.00 4.00 0.88
3 None - - - -0
4 AREA 2.00 7.00 4.00 0.88
5 AREA 2.00 7.00 4.00 0.88
6 VSAT 1.00 2.00 0.00 0.00
2.00 7.00 3.50 0.82
7 AREA+MOP
1.00 5.00 3.00 0.88
179
2.00 7.00 3.00 0.73

8 AREA+RAD
1.00 5.00 2.00 0.73
9 FXP 2.00 5.00 2.50 0.62
10 2.00 7.00 3.50 0.82
AREA+RAD
11 1.00 5.00 2.00 0.73
12 VSAT 1.00 2.00 2.00 0.73
13 MOP 1.00 5.00 3.00 0.88
- - - -
14 AREA+MOP
- - - -
- - - -
15 AREA+RAD
- - - -
2.00 7.00 4.00 0.88
16 AREA+RAD
1.00 5.00 2.00 0.73
17 MOP 1.00 5.00 2.00 0.73
18 FXP 2.00 5.00 2.00 0.50
2.00 7.00 3.00 0.73
19 AREA+FXP
2.00 5.00 2.50 0.62
20 MOP 1.00 5.00 2.50 0.82
21 MOP 1.00 5.00 4.00 0.95
22 AREA 2.00 7.00 4.00 0.88
23 None - - - 0.00
2.00 7.00 4.00 0.88
24 AREA+FXP
1.00 5.00 3.00 0.88
25 MOP 1.00 5.00 4.00 0.95
26 None - - - 0.00
2.00 7.00 4.00 0.88
27 AREA+MOP
1.00 5.00 3.00 0.88
2.00 7.00 4.00 0.88
28 AREA+FXP
2.00 5.00 3.00 0.73
2.00 7.00 4.00 0.88
29 AREA+FXP
2.00 5.00 3.00 0.73
30 MOP 1.00 5.00 3.00 0.88
31 FXP 2.00 5.00 3.00 0.73
2.00 7.00 3.50 0.82
32 AREA+MOP
1.00 5.00 2.00 0.73
180
Table 50 Mean certainty for each of the ICT deployments

ICT
ICT Deployment Max Score of Mean Certainty
0.73
VSAT 0.73
0.88
0.73
0.82
MOP 0.87
0.95
0.95
0.88
0.62
0.50
FXP 0.65
0.73
0.73
0.73
0.82
AREA+RAD 0.73 0.79
0.88
0.88
0.83
AREA+MOP 0.88 0.85
0.82
0.73
0.88
AREA+FXP 0.84
0.88
0.88
0.88
0.88
AREA 0.88 0.88
0.88
181
40.APPENDIX F: EFFECTIVENESS DATA FOR ALL THE CLIQUES
40.1.Full CAP Completeness
Table 51 Description of the ICT capabilities for the Full CAP Completeness set of parameters
Parameter AREA MOP RAD FXP VSAT
Ethnicity Text is only in MOP has the Text message Voice message Text message
English; audio on-board in English in any in English only
channel can J2ME applet only. FM radio language
carry a voice that can channel can
message in any display a SMS carry any
languages. text message language but
in all three HIH has no
languages. control over
the
commercial
FM channels
All-media All- Text message Fixed text Fixed text <description> <description>
hazards can carry indicating indicating element only element only
qualifier “warning” plus “warning” plus
elements of the limited length limited length
<alert> of the of the
segment, <description> <description>
<urgency>, element element
<severity>,
and
<certainty>
elements.
Audio message
can carry the
<description>
Multimedia Text and Text only Text only Audio only Text only
Audio
Table 52 Rating for the ICTs for effectiveness of relaying a Complete Full CAP message
AREA AREA AREA
Measure AREA RAD MOP FXP VSAT
+MOP +FXP +RAD
Ethnicity 1.00 0.15 1.00 1.00 0.15 1.00 1.00 1.00
All-media All-
0.95 0.70 0.70 0.80 0.70 0.95 0.95 0.95
hazards
Multimedia 0.90 0.50 0.50 0.80 0.50 0.90 0.90 0.90
Rating of Full CAP
0.86 0.05 0.35 0.64 0.05 0.86 0.86 0.86
Completeness
182
40.2.Active Alert Function
Table 53 Description of the ICT capabilities for the Active Alerting set of parameters
Parameter AREA MOP RAD FXP VSAT
Accountability Uses alternate Instantaneous Instantaneous Instantaneous Instantaneous
source to because a because a because voice when recipient
acknowledge return SMS is return SMS is call is real time clicks on
message sent to SMSC sent to SMSC and recipient is received
when message when message verified button TCP/IP
is viewed by is viewed by packet is
recipient after recipient after returned to
pressing button pressing button server
Wakeup Audible siren Audible siren Audible siren Audible Audible siren
on DAMB-R2 sound on hand and flashing ringing sound sound through
and external held light computer
siren speakers
Table 54 Rating of ICTs for effectiveness of Alerting the Last-Mile Communities

AREA AREA AREA
+MOP +FXP +RAD
Accountability 0.25 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Wakeup 0.87 0.95 0.87 0.85 0.70 0.87 0.87 0.95
Rating of Alerting
0.22 0.95 0.88 0.85 0.70 0.87 0.87 0.95
Functionality
40.3.Adoptability
The WorldSpace AREA and Speedcast VSAT systems required that the project invest in some of the
infrastructure such as purchasing a dedicated audio channel and installing satellite ground station for
the AREA and VSAT respectively. Although Table 14 shows the infrastructure cost these values are
not taken in to the calculation of the TCO.
Table 55 Costs associated with each of the ICTs

Units AREA MOP RAD FXP VSAT
Infrastructure Fixed Cost Per
0.00 0.00 0.00 0.00 20833.33
Cost Month
Infrastructure Variable Cost Per
1422000.00 0.00 0.00 0.00 90000.00
Cost Month
Terminal Device Fixed 8000.00 25000.00 15000.00 18000.00 100000.00
183
Cost
Terminal Device Cost Per
0.00 0.00 0.00 0.00 0.00
Variable Cost Month
Service Fixed Cost 0.00 1000.00 1000.00 0.00 50000.00
Service Device Variable Cost Per
1000.00 500.00 500.00 500.00 0.00
Cost Month
Number of Active Units Units 56 12 6 9 2
Total Cost for Terminal Cost Per
44000.00 44000.00 34000.00 36000.00 150000.00
Devices Month
Table 56 Total Cost of Ownership per house hold over three years
AREA AREA AREA
AREA MOP RAD FXP VSAT +MOP +RAD +FXP
TCO of deployed 80000.0
ICTs for 3 year 44000 44000 34000 36000 150000 88000 78000 0
TCO Per Family
for 3 years 440 440 340 360 1500 880 780 800.00
Effectiveness 1.00 1.00 1.00 1.00 0.00 0.79 0.58 0.62
Although the ICT Guardians were expected to document the daily journal it was not carried out.
Therefore, data on Integration of ICT in to daily life of the community is based on perception and a
few random interviews with the ICT owners. Researchers acknowledge that this data is not substantial
evidence.
Table 57 Description of the ICT integration in to daily life of community’s perceptions

AREA MOP RAD FXP VSAT
Enjoyed listening Monthly bills Not used at all Monthly bills Highly used by
to foreign NEWS show usage high negligible use of show usage high HIH for email,
channels such as during calamities FM radio. during calamities internet, and
BBC or NDTV as such as when the such as when the VoIP (Skype)
well as the Northeast conflict Northeast conflict
Sarvodaya Talk broke out in broke out in
channel January 2007. January 2007.
184
Table 58 Rating of ICTs for effectiveness of Adoptability

AREA AREA AREA
+MOP +FXP +RAD
Affordability 1.00 1.00 1.00 1.00 0.00 0.78 0.62 0.58
Utilization 0.70 0.10 0.95 0.95 0.95 0.95 0.95 0.70
Rating of
0.70 0.10 0.95 0.95 0.00 0.75 0.59 0.41
Adoptability
40.4.Miniaturization
Table 59 Weight of the ICTs in Kg

DAMB-R2, Nokia 6600 Remote 1xRTT CDMA
2.4m dish antenna,
Linear antenna Alarm Device 2000 telephone
iDirect modem, and
Power unit
0.1680 0.1250 3.6000 0.7500 44.6000
Note – The weight of the AREA+MOP, AREA+RAD, and AREA+FXP is the aggregate of the two
units
Table 60 Volume of the ICTs in cubic meters

Dimensions AREA MOP RAD FXP VSAT
Length DAMB-R2:
Modem + Power
0.1200 0.1086 0.275 0.0168
0.0508
Antenna
Width DAMB-R2: Modem + Power
0.0582 0.175 0.0189
0.0900 0.2889
Height DAMB-R2: Modem + Power
0.0237 0.0035 0.0077
0.0250 0.2413
Volume of
Cuboid 0.0003 0.0001 0.0017 0.0024 0.0035
objects
Radius - - - - 0.9000
Volume of
Spherical - - - - 1.2724
objects
Aggregated
0.0003 0.0001 0.0017 0.0024 1.2759
Volume
Note – The volume of the AREA+MOP, AREA+RAD, and AREA+FXP is the aggregate of the two
units
185
Table 61 Lifetime of Power of ICTs in hours

12v 7A Nokia 6600 in 12v 7A Inbuilt Prolink 0.6Kva
Rechargeable lithium battery Rechargeable rechargeable Uninterrupted
Matrix battery Matrix battery battery Power Service
8.00 36.00 4.00 8.00 0.5
Note – The lifetime of the AREA+MOP, AREA+RAD, and AREA+FXP is the aggregate of the two
units because one unit can be powered down when the other is operational.
Table 62 Rating of ICTs for effectiveness of Miniaturization

AREA AREA AREA
+MOP +FXP +RAD
Weight 1.00 1.00 1.00 1.00 0.29 1.00 1.00 1.00
Volume 1.00 1.00 1.00 1.00 0.10 1.00 1.00 1.00
longevity 0.50 0.25 1.00 0.50 0.03 1.00 0.75 1.00
Rating of
0.50 0.25 1.00 0.50 0.00 1.00 0.75 1.00
Miniaturization
40.5.Two-way
Table 63 Bi-directionality of ICTs in terms of upstream and downstream

Downstream only Upstream and Downstream and Upstream and Upstream and
(Broadcast) Downstream Restricted Downstream Downstream
Upstream
(Upstream works
only after alert
has been sent
and call back
number issued)
Table 64 Rating of ICTs for effectiveness of Two-way communication

AREA AREA AREA
+MOP +FXP +RAD
Rating of Two-way 0.00 0.70 1.00 1.00 1.00 1.00 1.00 0.70
186
41.APPENDIX G: OUTCOME OF THE ICT IN THE FIELD
41.1.WorldSpace Addressable Satellite Radio System
Reliability of transmitting CAP Messages
Initial test results show all devices to take less than 7.2 seconds to push the alert messages to the end
user AREA devices. The signal path for the alert message is from the Toulouse server to the Singapore
BOC, interfacing with the BC-LAN and inserting the binary coded CAP message into the OAAC
channel (BCID 2000) for decoding by the DAMB-R2. As DAMB-R2 was able to receive and interpret
the alert message, it appears that the CAP coding, binary conversion, routing to the correct port in
Singapore are happening correctly.
In terms of certainty there was no issue in receiving the minimum signal strength 4 of 7 bars to receive
text and audio alerts. The WS worked without fail in all of the coastal community live-exercises,
accept for one case when the antenna was not properly aligned for minimum required signal strength.
There was a situation where the CDMA and GSM terrestrial public networks work instructed to be
turned off by the Government of Sri Lanka in the North-East conflict areas during militaryy operations.
However, the satellite based one-way system was operational and was able to transmit alerts during the
field trials in these conflict areas. The average certainty of the WS AREA devices ranked to be 0.88
based on the average signal strength at each of the measuring points in the communities.
Efficiency as a function of time was measured from the instant the HIH staff created and issued the
CAP message through the ANNY Networks to the moment the message was received, decoded, and
recorded by the ICT-G in the community. In order to receive the text of all the elements, due to the
limited 80 character display, the user has to press one of the control buttons of the terminal device to
jump to the next text element. This step-by-step operation of scrolling and recording the text message
took 2 - 3 minutes. The HIH staff would spend 7 – 12 minutes, depending on the length of the
message to record the audio and upload to the server. Thereafter, the message would be queued
broadcasted within a few seconds. The WS system for the first set of live-exercises, with respect to a
tsunami alert anticipating a 90 minute period, ranked to have an efficiency of 0.87.
Effectiveness of the ‘Terminal Device
CAP Complete -- The WS additional ‘External Box’ could display the entire CAP message. The
‘External Box’ was not tested in the field (i.e. in the communities). The DAMB-R2 tested in the field
could display the CAP elements: <msgType>, <scope>, <sender>, , <status>, <category>, <urgency>,
<severity>, <certainty>, and <event>. These elements were adequate to provide the ICT-G with the
type of hazard and the priority level but inadequate to carry a complete alert bulletin with full
instructions. However, the audio message is capable of carrying the <description> element of the CAP
message. Both the text and audio components together form the mandatory elements of the CAP
Profile for Sri Lanka; as a result score a 0.95 on the ability to carry the necessary and sufficient
elements of a CAP message. The English text messages received in the rural English illiterate
communities were ineffective. Hence, an alternative was when issuing an alert using the ANNY
187
Network the HIH staff would instruct the interface to auto switch the audio channel of the DAMB-R2
to the SarvoTalk (BCID 950) channel. Thereafter, the HIH staff would record and upload an audio
(voiced) version of the <description> element of the CAP message, which would carry the complete
bulletin in the three national languages: Sinhala, Tamil, and English. The scoring method also
emphaisesd that the ICT be capable of transmit both audio and text, which gave the AREA a 0.95 score
for CAP Compliant (i.e. being CAP Complete).
Two-way – The WS systems being a one-to-multi-point system would score an absolute zero in terms
of effectiveness because the research emphasised the importance of tow-way communication in
warning systems. Therefore, the research had coupled the AREA units with an alternate two-way
device such as a GSM mobile phone or CDMA normadic phone. The ICT-G would use the mobile
phone or the normadic phone to communicate with the HIH to notify them of any local hazards
(upstream communication), which was not possible using the AREA units. This marriage would
increase the effectiveness score in terms of two-way of the AREA with mobile phone or AREA with
normadic phone to 0.85.
Adoptability – Total cost of ownership and integration of the system into the community daily life
measures the adoptability of the system. The cost of an AREA terminal device is well under USD 80.
However, the audio channel is what costs the most. Currently the HazInfo project is paying for the
audio channel. Sarvodaya being present in 15,000 villages in Sri Lanka can make it affordable to the
communities once the system is implemented in all those villages at which point the per user cost for
the Channel subscription fee will be affordable. One belief in the world of ‘public alerting’ is that the
best alerting device is the one that is ‘on’ and is ‘used’ all the time. The content for SarvoTalk (BCID
950) is developed by Sarvodaya. The channel is operational around-the-clock where Sarvodaya
broadcasts community-development based content to its member communities. There are a dozen
trained youth scattered in the coastal districts. They record content on to simple MP3 player/recorders
and email the files or deliver the files burned on CDs to the HIH. The HIH has employed a Media
Assistant who edits the content received from the communities and uploads on to the WS SarvoTalk
channel. Besides normal weekly programs the HIH also captures special events conducted by
Sarvodaya such as workshops and community events that are later broadcast to the communities,
which keeps the channel alive.
Miniaturization – The idea of the LM-HWS was to use the system during pre and post disasters; where
one aspect of the system is to provide alerts with adequate time for the communities to execute their
ERPs to protect themselves from the hazards and the second aspect is after the disaster when the
community livelihoods are destroyed the system must would communicate with the community during
response and recovery stages. Therefore, the terminal devices must be portable. The research assessed
Miniaturization in terms of the longevity of DC power consumption, volume of the complete terminal
device, and weight of the terminal device (including antenna and battery). The WS AREA-B system
can run on a 12v battery that weighs less than 3kg for 8 hours comfortably. The battery can be
recharged using a solar panel. The rating for the AREA unit (excluding the solar panel) based on the
fact that the unit must function consecutively for at least 30 hours, must weigh less than 30kg (for a
single person to carry comfortably), and has a volume that fits in a small suitcase or backpack, scored
0.68. The score improves to 0.95 when the portable solar panel is included in to the rating, which
assumes DC power is available for 18 hours (assuming 10 hours powered by solar panel during day
time and 8 hours of battery power during night time).
188
Alerting – This is an effectiveness feature equally important as CAP Completeness. The parameters:
Active Alert Function and Message Receipt Acknowledgement are what determine the Alerting
effectiveness. The message receipt acknowledgement is different from Two-way or upstream
communication. This function is important for the message ‘recipients’ to notify the message ‘sender’
the receipt of message [8]. Since WS uses forward-error-correction encryption and decryption for
audio and text packets it is almost impossible for the ICT-G to receive a mutated or truncated message,
provided the uplink and satellite doesn’t fail during transmission. The ICT-G would use the spouse
terminal device married to the AREA such as the GSM mobile phone or CDMA nomadic phone to
acknowledge message receipt. During a crisis situation both the GSM and CDMA networks get
congested and the acknowledgement would be delayed. Unlike an SMS controller that would receive a
return message upon the user accepting the SMS the WS system does not have a message returned to
the server, which the HIH staff could query to check if the messages have been delivered to the end
user. Active Alert Function describes the basis of being able to get the attention of the ICT-G. The
AREA units have an audible alarm. The research ranks an audible siren higher over a flashing light or
a vibration which must me in visible range or attached the ICT-G respectively [8]. Although the AREA
has the ability to ‘wakeup’ (or get the attention) of the ICT-G, it is weak in the message receipt
acknowledgement as it must rely on an alternate system. Therefore, the overall effectiveness for
Alerting scores a 0.22.
41.2.Dialog Disaster and Emergency Warning System
Uncertainty and Inefficiencies in Messaging
The terminal device reliability measures: certainty and efficiency were recorded on the day of the live-
exercises. The results in Fig. 5 are a summary (average) of the measures obtained during each of the
exercises conducted in the individual communities.
Certainty in the RAD devices dropped below a benchmark of 0.85 because the signal strength in
location of the devices measured between 2 – 3 bars. However, this did not effect the SMS alert
coming through because 1 bar was sufficient to receive the alerts and activate the alerting operational
states of the device. The research experienced cases of 0 (zero) certainty due to unquantifiable reasons
where the community ICT-G had accidentally deleted the J2ME applet, had forgotten to charge the
battery of the MOP, had not paid the phone bill, communities in the North and East conflict areas of
Sri Lanka being denied GSM access because the Government of Sri Lanka was conducting military
operations and had instructed the mobile operators turn of the cells in those conflict areas.
In some communities the alert message was not received during the first attempt of pushing the
messages to the MOPs using the DEWN Internet application. Therefore, the efficiency dropped as a
result of excess time taken to repeat the process in order for the MOPs to receive the alerts.
Normally both GSM devices take less than 30 seconds to push the alert messages to the end user
devices. No more that 1 – 3 units were used during any given trial. Therefore the trials were not
effected by congestion even thought both solutions use a store-and-forward method to push the SMS
189
alerts to each device. Ten MOPs were tested in the different communities. Hence, the additional
samples used in the trials compared to the 2 RAD units used the research had the opportunity to learn
lessons of unsuccessful behaviors. The MOPs tested in the rural environments, where the ICT
incompetence of the ICT-G in the communities using non traditional feature introduced by inserting a
J2ME applet, for example, further effected to the efficiencies because the menu and instructions were
in English, incomprehensible by folks in the rural areas. The RAD score a high rating in terms
efficiency because only 2 units were used in the field trials and they both happened to performed quite
well in the urban environment. The distinctly marked 4 buttons in he RAD opposed to the menu driven
MOP simplified the transition between operating states to increase efficiencies.
Effectiveness of Alerting the ICT-G
Both units, the RAD and MOP, have good active alerting capabilities because they both have audible
sirens and a flashing light. The MOP can be weak at times when the device is conceiled in a trouser
pocket or is ditached from the ICT-G at a distance; where both the audible siren and light may not be
strong enough to get immediate attention of the ICT-G. However, the RAD is the most effective unit of
all the ICTs tested. The RAD would gain the immediate attention of the ICT-G as a result of the loud
siren and externaly mounted large orange light even from a distance of 50 meters.
Of the two units the MOP has the best Miniaturization features; where the ICT-G can carry the MOP,
weighing less than 200 grams (Nokia 6600 weighs 130 grams), in an external belt-case or the pocket,
whcich also frees their hands. Comparitively, the weight and size of the RAD requires a single person
to carry the unit by its fixed handle or, as an alternative, carry the RAD in a backpack. The RAD works
both on AC and DC power. However, the 12v battery that provides the power to the unit had
approximately 4 hours or less of a life time, when the unit went through all the operational sates in
receiving multiplicity of repeated alerts . Unlike the MOP, which is quite versatile, the RAD does not
provide utilities other than the FM radio and for alerting for the average citizen to use in their daily
lives. The initial design of the units were for Government First-Responders; where the sets telephony
feature of the unit was denied. Therefore, the community users were not able to use the unit to make
voice calls or send SMS as an when they wanted, restricting upstream comunication. The voice call
was active as a ‘call-back” feature only following the receipt of and alert; where the user would press a
button to call back a central information source at a pre programmed phone number. The restriction in
two-way comunication in the RAD ranked very low compared to the MOP, which was unrestricted. In
regards to CAP Completeness, ony the MOP was capable of receiving messages in the three languages:
Sinahala, Tamil, and English. The RAD was capable of receiving Englishe text messages.
Shortcomings of the ICTs with the use of CAP
The 2 GSM ICTs used in the project make use of the CAP message format. As illustrated in Table 3,
each is only capable of displaying a limited number of the CAP elements, limiting the amount of alert
message content that can be transmitted to the ICT-G. Consequently, during simulations, ICT-G
recorded only the <msgType> and a truncated portion of the <description> elements because that’s all
they received. Although, the RAD and MOP devices use the <description> element that could have
carried a complete set of information, they were restricted to an overall message content limit of 270
190
characters, which is insufficient to carry a meaningful unambiguous alert message. Message did not
include <urgency>, <severity>, and <certainty> to enable ICT-Guardians to judge the Priority of the
incident, which is mandatory for the CAP Profile for Sri Lanka. Therefore, both devices were only
good for getting the attention of the ICT-G that there was an eminant threat but were forced to seek
complete information from an alternate source.
EXAMPLE 1: The simulations began at the HIH with the receipt of an email containing the critical
information pertaining to a Cyclone hazard. The following abstract of the entire message contains the
critical information of the message.
“A SEVERE CATEGORY 4 CYCLONE is now current for HAMBANTOTA District coastal areas.
At 10:00 am local time SEVERE TROPICAL CYCLONE MONTY was estimated to be 80 kilometers
west of Hambantota and moving southeast at 10 kilometers per hour.”
The HIH then transformed message to a CAP Message and entered the relevant information in to the
DEWN HTTP application. The <msgType> was set to “Warning” and the Cyclone message was
inserted in to the <description> field. The Cyclone message was translated to Sinhala and Tamil
languages to be used; especially with the MOP.
Table 3 illustrates the elements of the CAP message received and parts of the message received by
each of the devices. The RAD provided more information than the MOP but no a full message where it
was truncated after 160 characters. The MOP displayed that there was a Cyclone warning in effect but
had no additional information as to where and when the hazard is effective.
Table 65 Messages received by the RAD and MOP

RAD MOP
“Warning A SEVERE CATEGORY 4 CYCLONE “Warning”
is now current for HAMBANTOTA District coastal
areas. At 10:00 am local time SEVERE “A SEVERE CATEGORY 4 CYCLONE”
TROPICAL CYCLONE MONTY was estimated to
be 80 kilometers west of Hamban” “ {same in Sinhala}”
“ {same in Tamil”
41.3.Solana Networks/Innovative Technologies VSAT/IPAS
Reliability of the VSAT/IPAS
In tests performed with IPAS, the message system appears to be very efficient and straightforward for
composing and sending alerts. However, certain peculiarities were observed that will be important for
both HIH Monitor message senders and ICT Guardian receivers to take note of.
If the Internet connection is broken, the client may lose its subscription record and thus its
191
configuration for receiving alerts. When this occurs, the client still appears to the user to be functioning
correctly in its standby mode, but will not be triggered by an alert message until it is closed and
restarted. At that point, its subscription record is reloaded and the client will function correctly again.
However, even VSAT systems are not entirely risk free. Several factors can affect continuity of
services. First, the earth stations must be perfectly aligned with the geostationary satellite and any
small movement of the station can easily cause loss of signal. This can be problematic during severe
weather events such as cyclones, requiring skilled technicians to realign dishes and restore services.
Also, even when the satellite component of the network is functioning perfectly, loss of connectivity
from the gateway to the Internet can also disrupt connectivity for accessing the IPAS server or clients.
Such as the situation that occurred on 27 December 2006 when an earthquake in the Taiwan region
severed the undersea fiber network connecting the Hong Kong Gateway to Europe and beyond.
Further, any failure of equipment at the HIH or village nodes will cause message delivery failures.
Two potential recurring issues are power loss and affects of heat on satellite modems.
With the installation of the VSAT gateway at the HIH, two separate routes are now available to
improve redundancy and diversity in routing to the IPAS server located in Canada. One route (Dialog)
uses a domestic gateway to connect overseas with the return link for alerts traveling over the VSAT
stations fed through a gateway in Hong Kong. The second route uses the Hong Kong gateway to
connect the HIH to the Internet (in Hong Kong) and then is routed to the IPAS server whereupon the
return path from the server is uplinked to HIH and the VSAT equipped villages via the Hong Kong
gateway. In this latter case, the IPAS system can operate completely independently of the Sri Lankan
domestic networks and adds another potential layer of disaster resiliency.
Effectiveness of the System
Because IPAS, in its current form, is not CAP-compliant, it has not been tested for CAP
interoperability like the DEWN and ANNY applications, where a CAP message sent from the
WorldSpace ANNY application is teleported to the Dialog-Microimage DEWN application and,
thereafter, the DEWN application uploads the same message for dissemination. One advantage of
IPAS is that unlike SMS and the basic AREA DAMB-R2 components, IPAS appears capable of
carrying the equivalent of the entire CAP description field. IPAS alerts can also be received on any
Internet enabled MS Windows computer and is not dependent on a dedicated network. However, IPAS
currently only supports English language text messaging and cannot support Sinhala or Tami scripts in
either the Server GUI for message creation or Client GUI for message display.
The HIH depended exclusively upon a Dialog 128 kbps fixed terrestrial wireless link (bridge) for its
Internet access.13 This connection is terminated at the HIH through a modem which in turn is
connected to a Local Area Network (LAN) through a data switch and a wireless 802.11g WiFi access
point. Dialog provides 5 static IP addresses for use at the HIH.
Before the VSAT facilities were operational, this fixed wireless link was used to test the IPAS server-
to-client connection and to enable the HIH staff to become familiar with IPAS message composition
13
The Dialog connection is now replaced by the VSAT satellite connection and now serves as the backup connection.
192
and dissemination. However, during this period, a series of problems concerning outbound Internet
connectivity emerged and continued to plaque the tests, resulting in both the software timing out before
messages could be fully entered into their templates and/or sent. These problems manifested
themselves in the form of severe network congestion that caused lengthy delays in logging into the
servers. After an extensive investigation, it was revealed that four separate issues were at the root of
the problems.
The first problem was caused by an incorrectly connected cable that was routing data packets from the
WiFi access point router back into the main Dialog connection. This effectively caused the LAN to be
flooded with excess packets, causing severe local congestion. Once correctly cabled, most of the
congestion disappeared.
The second problem was caused by the HIH workstations being connected to the WiFi access point,
rather than through a cable connection to the Dialog router via the data switch. The WiFi access point
appears to intermittently drop its signal, causing the link to time out and PCs to have to reinitialize
their Internet connections through a DHCP setup. Each time this occurred, the connection to the IPAS
servers broke and/or timed out. This problem was solved by connecting the PCs directly to the Dialog
connection and assigning them a Dialog static IP address.
The third problem was caused by congestion on the main Dialog backbone connection. This was
discovered by conducting a trace route connection between the HIH and the final destination where the
server resided; in the case of IPAS, the server resides Canada. A trace route test sends a series of data
packets through the network and measures and analyzes the quality of the link according to how long it
takes to get an acknowledgement back from all of the locations where the packets are routed on their
way to the final destination (latency), along with how many of the sent packets were dropped in transit
(packet loss). The trace route tests revealed that, at times, there were significant latencies in packet
delivery possibly due to congestion or a malfunctioning router at the Dialog gateway, which were
resulting in timeouts and excessive packet losses en route to their final destination. Results of the route
traces were sent to Dialog who, in turn, made changes to the packet routing. The routing now appears
to be stable.
The fourth problem was caused by the Dialog radio antenna being mounted on a tall pipe that tends to
sway in the wind, moving the antenna in and out of focus.
Power outages and lightning storms -- VSATs and PCs require an alternative power source during
power interruptions. Presently, neither the HIH nor the ICT Guardians possess alternative backup
power to mitigate the impacts of frequent or prolonged power interruptions. Often, to prevent power
surges, equipment is unplugged during lightning storms.
Further in order to keep satellite modems functioning properly they must be protected from excess heat
and require proper ventilation and some form of air conditioning. Even accidentally covering the
ventilation holes on the modem’s cover can cause equipment malfunctions.
193
42. APPENDIX H – DATA FROM SILENT TESTS
Table 66 Silent test results 06 to 10 August 2007

No Date Time Village Device No. ICT Status Remarks
1 Moratuwa 602164957 FXP Working
2 06.08.2007 12.03PM Thirukkadalar 602269982 FXP Working Line is Not Clear
3 06.08.2007 12.07PM Oluvil 602679957 FXP Working
4 10.08.2007 9.23AM Samudragama 602269981 FXP Not working Phone is Currently Not Available
5 10.08.2007 9.26AM Shithtandikudipuram 602659950 FXP Not working Phone is Currently Not Available
6 06.08.2007 1.10PM Waththegama 602418972 FXP Working
7 06.08.2007 12.50PM Kottegoda 602418973 FXP Not working Ringing NO Answer
8 06.08.2007 12.54PM Idivinna 602489978 FXP working
9 Periyakallar 602659955 FXP Not working
10 Oluvil 602679957 FXP Working Now the Line is Clear
11 07.08.2007 10.08AM Palmunnai 773428477 MOP Working
12 06.08.2007 4.28PM Welhengoda 773428478 MOP Working
13 10.08.2007 3.42PM Thalalla South 773428479 MOP Working DEWN software is not working
14 10.08.2007 3.37PM Meddewatte 773428480 MOP Working
15 06.08.2007 4.38PM Saturkondan 773428481 MOP Not working Line is Disconnected
16 06.08.2007 2.22PM Diyalagoda 773428482 MOP Working
17 10.08.2007 3.22PM Thambiluvil 773428483 MOP Working
Niluka
18 10.08.2007 3.45Pm Wicramasinhge 773428484 MOP Working
19 06.08.2007 4.25PM Nindaur 773428485 MOP Working
20 07.08.2007 10.23AM HUB Disconnected 773428486 MOP Not working Line is Temporally Disconnected
21 07.08.2007 3.01PM Samodagama 773428487 MOP Working DEWN Software is Uninstalled
22 Rajiw 773428488 MOP Working
10000000
23 Moratuwella 00068845 AREA
194
10000000
24 Maggona 00068841 AREA
10000000
25 07.08.2007 4.10PM Palamunai 00068842 AREA
10000000
26 07.08.2007 4.22PM Warregama South 00068840 AREA
10000000
27 Brahmanawatta 00068847 AREA
10000000
28 Kalmunai II 00068849 AREA
10000000
29 06.08.2007 4.28PM Wellhengoda 00068848 AREA
10000000
30 Susiripala 00068843 AREA
10000000
31 Talalla South 00068844 AREA
10000000
32 Wennamulla 00068800 AREA
10000000
33 Panama North 00068805 AREA
10000000
34 Sithandikudipuram 00068801 AREA
10000000
35 Modara Pelessa 00068806 AREA
10000000
36 Munnai 00068802 AREA
10000000
37 Tondamanaru 00068807 AREA
195
Table 67 Silent test results 15 to 20 October 2007

No Date Time Village Device No ICT Status Remarks
1 16.10.2007 9.10AM Moratuwa 602164957 FXP Working Some times it is not clear
2 16.10.2007 9.22AM Thirukkadalar 602269982 FXP Working
3 16.10.2007 9.25AM Oluvil 602679957 FXP Working
could not do the silent test because
4 16.10.2007 9.26Am Samudragama 602269981 FXP Not working phone is not contactable
Could not do the Silent test
5 16.10.2007 9.34AM Shithtandikudipuram 602659950 FXP Not working because the phone is Power off
Could not do the silent test because
6 18.10.2007 9.46AM Waththegama 602418972 FXP Not working phone is not reachable
7 17.10.2007 9.51AM Kottegoda 602418973 FXP Not working phone is not contactable
8 18.10.2007 9.55AM Idivinna 602489978 FXP Working
9 17.10.2007 10.04AM Periyakallar 602659955 FXP Ringing Phone is Ringing but no answer
10 17.10.2007 0 Palmunnai 773428477 MOP Working
AREA is working but could not do
the silent test because the person is
on the way to Sarvodaya Head
11 16.10.2007 10.04Am Welhengoda 773428478 MOP Working office
12 20.10.2007 4.28Pm Thalalla South 773428479 MOP Working DEWN software is not working
13 20.10.2007 3.42PM Meddewatte 773428480 MOP Working
15 16.10.2007 4.38PM Diyalagoda 773428482 MOP Working
16 20.10.2007 2.22PM Thambiluvil 773428483 MOP Working
Niluka
17 20.10.2007 3.22PM Wicramasinhge 773428484 MOP Working
19 17.10.2007 4.26PM HUB 773428486 MOP Not working Phone is Temporally Disconnected
20 17.10.2007 10.23AM Samodagama 773428487 MOP Working DEWN Software is Uninstalled
21 18.10.2017 10.30AM Rajiw 773428488 MOP Working
196
The ICT person's all of contact lines

10000000 are not working. Could not do the
22 15.10.2007 11.10Am Moratuwella 00068845 AREA - silent test
it is a Religion problem she is
handed over the AREA Device to
Temple through the Sarvodaya
District Center. Have no any
10000000 contact details of the temple
23 15.10.2007 11.35Am Maggona 00068841 AREA - persons.
24 15.10.2007 11.42AM Palamunai 00068842 AREA - silent test
The ICT person's all of contact
10000000 lines are not working .Could not do
25 18.10.2007 9.46AM Warregama South 00068840 AREA - the silent test
The Power charger is not inside.
10000000 After the rapier the power charger
26 18.10.2007 10.01AM Brahmanawatta 00068847 AREA - is misplaced
10000000
27 18.10.2007 10.11AM Kalmunai II 00068849 AREA - could Not contact to person
10000000
27 18.10.2007 10.15Am Wellhengoda 00068848 AREA Working
29 18.10.2007 10.17AM Susiripala 00068843 AREA - silent test
10000000
30 18.10.2007 10.20Am Talalla South 00068844 AREA Working Some times it is not working
31 18.10.2007 10.49Am Wennamulla 00068800 AREA - silent test
32 18.10.2007 11.07AM Panama North 10000000 AREA - The ICT person's all of contact lines
197
are not working. Could not do the

00068805 silent test
33 18.102007 11.10Am Sithandikudipuram 00068801 AREA - silent test
34 18.102007 11.18Am Modara Pelessa 00068806 AREA - silent test
10000000
34 - - Munnai 00068802 AREA -
10000000
36 - - Tondamanaru 00068807 AREA -
Table 68 Silent test results 26 to 30 December 2007

No Date Time Village Device No ICT Status Remarks
1 26.12.2007 11.16AM Moratuwa 602164957 FXP Working
2 26.12.2007 11.32AM Thirukkadalar 602269982 FXP Working
3 26.12.2007 11.38AM Oluvil 602679957 FXP Working
Power OFF ; could not do the silent
4 26.12.2007 11.38.AM Samudragama 602269981 FXP Not working test because phone is Power Off
Power OFF ; Could not do the
Silent test because the phone is
5 26.12.2007 11.40AM Shithtandikudipuram 602659950 FXP Not working Power off
Power OFF ; Could not do the
silent test because phone is not
6 26.12.2007 12.00.PM Waththegama 602418972 FXP Not working reachable
7 26.12.2007 11.48AM Kottegoda 602418973 FXP Not working phone is not contactable
8 26.12.2007 11.45AM Idivinna 602489978 FXP Working
9 26.12.2007 11.42AM Periyakallar 602659955 FXP Ringing No answer
198
10 26.12.2007 11.46AM Palmunnai 773428477 MOP Working

11 26.12.2007 12.25PM Welhengoda 773428478 MOP Working
12 26.12.2007 1.03PM Thalalla South 773428479 MOP Working
14 26.12.2007 1.07PM Diyalagoda 773428482 MOP Ringing Not Responded
15 26.12.2007 1.08PM Thambiluvil 773428483 MOP Not working Out of Service; disconnected
Niluka
16 26.12.2007 11.17AM Wicramasinhge 773428484 MOP Working
18 26.12.2007 11.18AM HUB Disconnected 773428486 MOP Not working Phone line is Disconnected
19 26.12.2007 11.18AM Rajiw 773428488 MOP Working
10000000 lines are not working. Could not do
20 30.12.2007 9.48Am Moratuwella 00068845 AREA Not working the silent test
10000000 The ICT Person's Contact line is
21 30.12.2007 9.49AM Maggona 00068841 AREA Not working can not be reach
22 30.12.2007 9.50Am Palamunai 00068842 AREA working the silent test
10000000 lines are not working .Could not do
23 30.12.2007 12.25PM Warregama South 00068840 AREA Ringing the silent test
The Power charger is not available.
10000000 Other After the repair MOP the power
24 26.12.2007 Brahmanawatta 00068847 AREA Problem charger is misplaced
10000000
25 12.25PM Kalmunai II 00068849 AREA - could Not contact to person
10000000
26 30.12.2007 12.27PM Wellhengoda 00068848 AREA Working
27 30.12.2007 1.03PM Susiripala 10000000 AREA - The ICT person's all of contact
199
the silent test

10000000
28 26.12.2007 1.06PM Talalla South 00068844 AREA Working Some times the signal is problem
29 26.12.2007 9.51Pm Wennamulla 00068800 AREA - the silent test
Could Not contact the person; The
ICT person's all of contact lines are
10000000 not working. Could not do the
30 30.12.2007 9.53PM Panama North 00068805 AREA - silent test
Can not contact the person; The
10000000 ICT person's all of contact line are
31 30.12.2007 9.54PM Sithandikudipuram 00068801 AREA - switch off
Power off; The ICT Guidant Is no
at the home. Spoke with his wife
10000000 she said that the Device is not
32 30.12.2007 12.01PM Modara Pelessa 00068806 AREA - connect t the Power.
10000000
33 - Munnai 00068802 AREA - -
10000000
34 - Tondamanaru 00068807 AREA - _
200
43.APPENDIX I – OUTCOME OF ICT ON 12-SEP-2007 TSUNAMI ADVISORY
Table 69 Email from Executive Director of Sarvodaya on Tsunami Watch 12-Sep-2007

Dear All,
It was quite an interesting learning experience how the country (Sri Lanka) reacted to yesterday's
Tsunami warning following the earthquake in Indonesia yesterday and also the present alert following
another quake around 5.30 a.m. this morning.
Sarvodaya Community Disaster Management Centre (SCDMC) went into immediate action and we
were able to test "live" all our systems including the landlines, mobile phones and especially the
addressable satellite radio (ASR). After the official Tsunami warning was given we prepared and sent
regular alerts and updates to the Last Mile Project (IDRC supported) villages as well as the Sarvodaya
District Centers in the coastal belt. We are receiving now reports of how effectively the messages were
received.
We were also able to identify weaknesses in our own systems at the Emergency Operations Centre
(EOC) of SCDMC and today as I write this another Tsunami warning has been given and we are
working on these weaknesses.
We have also witnessed how much the national disaster warning system has to improve in terms of
standard protocols etc. There was wide spread panic and uncoordinated response.
This is only an initial update for our partners and a detailed report will be submitted towards this
evening.
Regards,
Vinya
Table 70 Performs of Mobile Phones on the real event of a Tsunami watch on 12-Sep-2007
0 ICT G Name Tel No Message Status Remarks
1 Hewawitharana 773428480 Not Received Software Problem
2 Ranjith 773428481 Not Received Service Disconnected
0 Salee 773428482 Not Received Problem with provider service
3 P.Alahuthurai 773428483 Received
4 HIH 773428484 Received
0 Rafeek 773428485 Not Received Problem with provider service
5 HIH 773428486 Not Received Line is Disconnected
6 J.A.C.Priyadarshani 773428487 Received Message received but she couldn't
understand message because the
201
mobile software is removed
0 Not Nominated 773428488 Received

7 A.M.Yasar 773428477 Received
8 K.Nishantha 773428478 Not received Problem with provider service
0 Ajith Soyza 773428479 Received Software Problem; Message received
but couldn’t understand because
displayed only the DEWN software
name
9 HIH Back up 773428476 Received
10 HIH Back Up 773428475 Received
202
44.APPENDIX J – DATA COLLECTION SHEETS
44.1.Datasheet 1 – ICT Guardian Alert Log
Last Mile Hazard Warning System LMHWS Datasheet

001
Last mile - 
  
,Wjpf; fl;l mdu;j;jj;ij vr;rupf;fpd;w mwptpj;jy; rhjdk;
ALERT LOG
.u osi;
a %l
s l
a h
fpuhkk; khtl;lk;
Village District
,enqKq oskh ksl=;a l< oskh iy

iy fõ,dj fõ,dj
fpilj;j ntspapl;l NeuKk;
NeuKk; jpfjpAk;
jpfjpAk; Disseminate Date Time
Received
DateTime
lreKdlr my; i|yka WmlrK w;ßka i¾fjdaoh wdmod f;dr;=re uOHia:dkfhka Tng ,enqKq NdKavh jgd rjqula w¢kak –
jaT nra;J fPo;f;Fwpg;gpl;Ls;s cgfuzq;fspy; ru;Nthja mdu;j;j jfty; epiyaj;jhy; toq;fg;gl;l fUtpia Rw;wp
tl;lkpl;Lf; fhl;Lf.
Please circle the ICT which you operate to receive Alerts from the Sarvodaya HIH.
ia:dk.; cx.u ÿrl:kh ù-ieÜ wdmod {d

ÿrl:kh ifalf;fj; tP-nrl; ikaksfõok WmlrK
epiyahd njhiyNgrp VSAT .=jkaúÿ,sh njhiy
njhiyNgrp Java Phone nra;kjp thndhyp ,af;fg;ghl;L
Fixed Line AREA vr;rupg;Gf; fUtp
203
Remote Alarm
Device
ix{d ^ mrSlI
a K fyda WoajSmk & my; j.=fjs igyka l, hq;= w;r Tn .kq ,enQ l%shd udra.h o igyka
lrkak
fpuhkj;jpw;F fpilf;fg; ngWk; vitNaDk; guPl;rhj;jpa my;yJ cz;ikahd vr;rupf;iffs; njhlu;ghd tptuj;ij
kw;Wk; re;ju;g;gj;ij gad;gLj;jpa tpjk; njhlu;ghf fPo;fhZk; gl;baiy )u;j;jp nra;J mDg;gpitf;fTk;
All alert is received, including tests and exercises, the Village ICT Guardian must record the Alert and describe the action
taken.
wx.h w.h wx.h w.h

%yk; ngWkjp %yk; ngWkjp
ix{dj NdIdj:
vr;rupf;if nkhop
Alert Language
yÿQkAjkh
j¾.h
milahsg;g
tif
Lj;Jdu;
Category
Identifier
hjkakd isoaêh
mDg;Ggtu; rk;gtk;
Sender Event
heõfõ yÈish
mDg;gpaJ mtruk;
Sent Urgency
;;Ajh ±âlu
epiyik Jbdk;
Status Severity
mKsúvfha
iA:srNdjh
iajNdjh
epiyahd
jftypd;
jd;ik
jd;ik
Certainty
msgType
204
ljqreka
ioydo
Jiw
Scope
úia;rh
tptuk;
Description

jfty;
%yq;fs;
Resource
m%foaYh
gpuNjrk;
Area
.kq ,enq l%shdu¾.h
205
Nkw;nfhz;l nraw;ghL
Action Taken by ICT Guardian
ix{d Ynsoh ,enqfkao ?

/  / Did
you hear the alerting
sound? Yes/no and
comments.
ix{dj f;areus .ekSug

myiqo ?
/  / Was
the alert message easy to
understand? Yes/no and
comments.
?
/ 
/ Was the ICT easy to use?
206
Yes/No and comments.


/ 
/ Do you feel you have sufficient training to operate
the ICT? Yes/No and comments.

gpur;rpidfSk; fUj;Jf;fSk;
Comments
WmlrK Ndrlref.a ku/ cgfuz w;aik / ifnahg;gk; Èkh /jpfjp

nghWg;ghsupd; ngau; Signature Date
/ICT Guardian
207
44.2.Datasheet 3 – Community drills evaluation form
Page 1 Last Mile Hazard Warning System LMHWS Datasheet

.ïudk fmryqre jevigygyk-m%yYakdj,sh iy idrdxYh / 003
,Wjpf;fl;l mdu;j;j vr;rupf;if xOq;F Kiw /fpuhkpa
xj;jpif kjpg;gPl;L-rhuhk;rKk; tpdhg;gj;jpuKk;
Village Simulation Count-Summary and Questionnaire
.u / fpuhkk; / Village osi;

a %l a h / khtl;lk; / District
s l
oskh /jpfjp / Date fõ,dj / Neuk; / Time
Please fill in the answers to the questions by asking the questions from the village simulation participants that are at the
final assembly point.
ldKavh/ tu;f;fk; / Category msßñ / Mz; / Male .eyeKq / ngz; / Female <uhs /rpWtu; / Children
.ïjdiskaf.a iyNd.s;ajh/
vz;zpf;if / Number of
Community Participants
.ïjdis mjq,a ixLHdj / FLk;g vz;zpf;if / Number of Families in Community

.fus uqM ck.ykh/fpuhkj;jpd; nkhj;j rdj;njhif / Total Community Population
208
Q1
(a) .ïjdiska fldmuK ixLHdjla bj;ajSfñ ie,iqu ms,snoj ±kqj;aj isáfhao? / mtrufhy ntspNaw;wy;
jpl;lk; njhlu;ghf vj;jid Ngu;fs; mwpe;Js;sdu; ? / How many knew the emergency evacua-
tion plan for the community?
Q2 (a) .ïjdiska w;r mj;sk wk;=re we.ùfñ l%u fudkjdo?/ vit cs;@u; vr;rupg;Gfshf ,Ue;jd ?
/ What
(b) were the
.ïjdiska local warning
fldmuK ixLHdjlamethods?
bj;aùfñ ud¾.h ms,snoj wlue;a; m,lf,ao?/ vj;jid Ngu; ntspNaw;wy;
ghijfs; gw;wp Fyg;gkile;jtu;fshf ,Ue;jdu; ? / How many were confused over evacuation
routes?
(b) tu l%u ls%hd;aul lsÍug .;jk ld,h?/ cs;@u; vr;rupg;G fUtpfis ,af;Ftjw;F vt;tsT
Neuk; vLj;jJ ? / How long did it take to activate the local warning systems?
(c) iEu wdmodjla i|ydu bj;ajhdfñ ie,iqula Tfì .ïjdiska i;=j ;snqfkao? (Tú /ke; / midj;J
mdu;j;jq;fSf;Fk; Kfk; nfhLf;ff;$ba jpl;lk; xd;W ,Uf;fpd;wjh ?
(c) wk;=re we.úfñ ix{dj weiqkq .ïjdiska ixLHdj ?/ vj;jid Ngu;fs; vr;rupg;G rkpf;iQia
/ Does your community have a plan for all-hazards (yes/no and reasons)?
mtjhdpj;jdu; ?/ How many heard the alarm?
(d) bj;a fkdjqkq .ïjdiska msßila isáfhao? / ntspNaw;wypd;NghJ fpuhkj;jpy; VjhtnjhU gpupT
(d) ix{dj weiqkq .ïjdiska u.ska ±kqj;a jqkq .ïjdiska ixLHdj ? / vj;jidg; Ngu;fs; vr;rupg;G
tpLg;gl;ljh ? / Were any parts of the village not evacuated?
rkpf;iQia Nfl;ltu;fshy; mwpTWj;jg; gl;ldu;? / How many were notified by the people who
heard the alarm?
(e) bj;aùfñ ud¾.h yryd ndOl lsisjla ;snqfkao? (Wod:- ÿñßh yria ud¾., md,ñ, lgq lñì jeg,
bj;al, fkdyels WmlrK) / ntspNaw;wy; ghijapy; VjhtJ jilfs; ,Ue;jdth ?
(e) wk;=re we.úfñ ix{dj u.ska wk;=r meyeÈ,sj yÿkd.; yels ùo? bkamiq ta ioyd ls%hd;aul jSug
(ghyq;fs;>Gifapuj flitfs>fjTfs;> Kw;fk;gp Ntspfs; ) Nghd;w / Were
fldmuK ld,hla ;snqfkao? / vr;rupf;if muptpg;G mghaj;ij NtWgLj;jp mwpe;J r
there any obstructions, stored materials, equipment blocking the evacuation paths, bridges,
%fj;jpw;F cupa mwptpg;ig toq;f Kbe;jjh ? Mk; >,y;iy vdpd; fhuzk;
railway-crossings, doors, barbwire fences, etc, )
/ Did the alarm distinctly identify the hazard? And the available time for the community to react (yes/no
and reasons)?
(f) wk;=re we.úfñ ix{dj kdoùfuka wk;=rej wdrCIs; iaMdkh lrd hdug .;jk ld,h? /Kjy;
vr;rupf;ifapypUe;J ghJfhg;ghd ,lj;jpw;F ntspNaww;g; gLk; tiuf;Fk;
vt;tsT Neuk; vLj;jJ ?How long did it take from first alarm to evacuate or move to
shelter area?
Q3 (a) .ïjdiska fldmuK ixLHdjla bj;aùfñ fmryqrejg iyNd.s ùug wlue;a; m,lf,ao ?/ jid Ngu;fSf;F
,e;j xj;jpifapy; fye;Jnfhs;s tpUg;gk; ,y;iy ? / How many did not want to participate in
the Simulations?
(b) ikaksfúok WmlrK fol ne.ska ,o .ïjdiska fldmuK ixLHdjla bj;aùfñ fmryqrejg iyNd.sùo ? /
209
vj;jid Ngu; vspjpy; efuf;$ba – gytPdg;gl;Nlhu;fs; tUifj; je;jpUe;jhu;fs; ?

/ How many mobility-in paired people present?
(c) bj;aùfñ fmryqrejg iyNd.s fkdjSug fya;=?/ ,e;j xj;jpifapy; fye;Jnfhs;shjikf;F

fhuzj;ij fhl;Lf.? / What are the reasons for the people not to participating in the Simulation?
(d) .ñudkfha iEu m%foaYhlgu wk;=re we.ùfñ mKsúvh m%pdrKh ùo?/ cs;@u; mghaj;ij r
%fj;jpdhy; vy;yh ,lj;jpYk; Nfl;f Kbe;jjh ? / Were local alarms for warning, heard from all
occupied areas of the community?
HIH bÈß ie,iqñ flfia

(e) wdmod ioyd iqodkñ ùfñ lghq;= flfrys .ïjdiskaf.a iyNd.S;ajh jeä lr.ekSug
úh hq;=o? / mdu;j;jj;ij vjpu;nfhs;tjw;fhd jahu;g;gLj;Jjy; r%f gq;fspg;ig
Kd;Ndw;w HIH jpl;lk; vt;thW cjt KbAk; ?/ How can the HIH plan in the future to improve
the community participation in disaster preparedness?
Q4 (a) fuu jevigyk id¾:l f,i ms,s.;a ixnHdj? /vj;jid fw;gid rk;gtq;fs; ntw;wpf;nfhz;ld ?
How many thought the event was a success?
(b) fuu jevigyfkys ÿ¾j,;d rdYshla we;s nj;a ;j ÿrg;a fuu wdmod ikaksfúok ie,iqu ioyd iydh
wjYH f,I is;= ixnHdj? /mtrufhyepiyf;F Kfk; nfhLg;gjw;fhd jpl;lj;ij tpUj;jp
nra;a NkYk; cjtp Njit vd;Wk; mjpy; gy gpur;rpidfs; cs;sd vd;Wk;
vj;jid Ngu; epidj;jhu;fs; ?How many thought there were many problems and the village needed
further assistance in developing an emergency response plan?
(c) .ïjdiska ioyd wdmod l,uKdlrkh iñnkaoj ixúOdk iydh, ;dCIKsl iydh iy ICT Ndú;h iñnkaoj
iydh ;j ÿrg;a wjYHo? /ICT ia gad;gLj;JtjpYk; mtrufhy jpl;lj;jpYk; xOq;FgLj;jy;>
njhopy; El;gk; vd;gtw;wpw;F Nkyjpf cjtp Njitah ? Does the community need
additional help in organizational and/or technical areas of emergency planning and ICT usage?
210
Q5 (d) ICT WmlrK ksis wdldrfhka ls%hdlrkafkao? ls%hdfkdlrkafka kñ Bg fya;=?

Kd;ndr;rupg;G cgfuz ,af;Fjy; njhlu;ghd tpsf;fk; nfhz;Ls;sPuh ? ,y;iyNay;
mjw;fhd fhuzk; vd;d ? Did the community ICT for early warnings function properly? If no what
were the faults?
(e)jevigyfkys fjk;a fodaI?/NtW VjhtJ FiwghLfs; my;yJ gpiofis

mtjhdpf;fg;l;ldth ? ( Mk; ,y;iy ) c+k; : nrad;Kiw Nghjhik> kWnkhop
toq;Fk; Neuk; mjpfk;> mwptpg;ig Nfl;f KbahJs;sik )Were any other faults or
deficiencies noted Yes/No; (e.g. procedures inadequate, response time too long, announcements inaudible
etc)
fjk;a úfYaI lrekq we;akñ my; fldgqfjys ioyka lrkak:Use space below for additional remarks: fPo; fz;l
,ilntspapy; Fwpg;Gfs; vOJtjw;F gad; gLj;Jf.
m%dfoaYSh Èia;s%la kshuq w;aik / ifnahg;gk; Èkh /jpfjp

fiajl/gpuNjr ,izg;ghsu; / Divisional Signature Date
211
Coordinator
212
44.3.Datasheet 4 – Counting people at strategic points

reiapk ia:dkfha f;dr;=re / ,Wjp mdu;j;j 004
vr;rupf;if xOq;F Kiw -xd;W$Lk; vz;zpf;if
Assemble Point Counts
.u reiapk ia:dkh .eyeKq / msßñ / <uhs

fpuhkk; xd;W$Lk; ,lk; Mz;/ngz;/rpWtu;
Village Assemble Point Male/ Female/ Children
oskh ksl=;a l< oskh

Muk;g iy fõ,dj
jpfjpAk; KbT Neuk;
NeuKk; Finish Time
Start Date
and Time
•iEu ñks;a;= 15lg jrla wod, fldgqfõ l;sr ,l=K oukak' / xt;nthU 15 tJ epkplj;jpYk; Fwpj;j rJuj;jpy; Fwpj;j Neuj;ij ,lTk; / Every
15 minutes record the time in the last box crossed
• by, jñ fl,pf¾ fldgqfjka wdrñN lr, .Kkh lrk iaMdkh yer hk iEu wfhla fjkqfjkau l;sr i,l=Kq lrkak
• cq;fs; kjpg;gPl;L ];jsj;jpypUe;J flf;Fk; xt;nthU egufspdJk; Fwpg;GfSf;fhf rJug;ngl;bfspy; Nky; ,lg;gf;fkpUe;J Gs;sb ,lTk;.
• Cross out the boxes starting from upper left corner working each row for every person that passed your counting station
213
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80
81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120
121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140
141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160
161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180
181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
……………………….. ………………… ………………………………..

n,Odßhd / cupikf;fhuu ;/ Authorized Person w;aik / ifnahg;gk; / Signature
214
44.4.Datasheet 5 – ICT Guardian daily Test Log
Last Mile Hazard Warning System

Last mile - 
DAILY TEST LOG
LM-HWS Datasheet 5
 
 
 
  
Village District Month


Please circle the ICT used for early warnings through the Sarvodaya Hazard Information Hub –
    

Fixed Line Java Phone VSAT AREA RAD

Time
     
ICT Guardian Name Day AM PM Comments Signature
1
2
3
4
5
6
7
8
9
10
11
12
13
215
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
District Coordinator Name Signature Date

  
216
44.5.Datasheet 6 – ICT Guardian Breakdown Log

ls%hdúrys; ùfñ f,d.h/ ,Wjpf;fl;l 006
mdu;j;j vr;rupf;if xOq;F Kiw
BREAKDOWN LOG – nrayw;w
rhjdk;
.u osi;
a %l
s l
a h udih
fpuhkk; khtl;lk; khjk;
Village District Month
lreKdlr my; i|yka WmlrK w;ßka i¾fjdaoh wdmod f;dr;=re uOHia:dkfhka Tng ,enqKq NdKavh jgd rjqula w¢kak –
jaT nra;J fPo; Fwpg;gpl;Ls;s rhjdq;fspy; ru;Nthja mdu;j;j jfty; epiyaj;jhy; toq;fg;gl;l fUtpia
tl;lkpl;Lf; fhl;Lf.
Please circle the ICT which you operate to receive Alerts from the Sarvodaya HIH.
{d
ia:dk.; wdmod WmlrK
cx.u ÿrl;kh
ÿrl;kh ù-ieÜ ikaksfõok .=jka njhiy
ifalf;fj;
epiyahd tP-nrl; úÿ,sh ,af;fg;ghl;L
njhiy Ngrp
njhiyNgrp VSAT nra;kjp thndhyp vr;rupg;Gf; fUtp
Java Phone
Fixed Line AREA Remote Alarm
Device
WmlrK Ndrlref.a ku
cgfuz nghWg;ghsu; ngau;
ICT Guardian’s Name
217
Èkh iy úia;rh
fõ,dj tptuk;
jpfjpAk; Description
NeuKk;
Time and Date
fodaIh
NfhshU/Njhrk;
Problem
yels kï fodaIh ms<sn|

úia;rhla imhkak
KbAkhapd; Nfhshupd;
jd;ikia Fwpg;gpLf
Description of the problem
Èkh iy
fõ,dj jd¾:dj ,nd .;af;a
jpfjpAk; Tjk wdlrh mwpf;ifia ngw;Wf;
NeuKk; mDg;Gk; tpjk; nfhz;ltu;
(fax, phone, email, sms, etc) (HIH person name):
jtW/gpur;rpid/gpiof;Fwpa
mwpf;if mDg;gg;gl;lJ
Fault/Problem/Error Report
Sent
ngwg;gl;lj; jPu;Tfs;
Solution Received
218
If the solution was from a source other than the HIH, then specify source and comment on how you got the external help in
the additional information section below; or you may use the area to provide any other relevant information.
jeä õia;r
Nkyjpf jfty;fs;
Comments :
Èia;%sla kshuq fiajlf.a ku w;aik Èkh

khtl;l ,izg;ghsuJ ngau; ifnahg;gk; jpfjp
District Coordinator Signature Date
219
44.6.Datasheet 7– Event of Interest form
220
44.7.Datasheet 8 – HIH Monitor drill performance evaluation
 



Observer Name,
Occupation, Affiliation,
designation
 
Date: Time:

 

Write the name, role, and designation of HIH resource person particiapting in the last-mile HazInfo Project simulation
activities.
   

Resource Person Designation Role Other Details
A
221
 
Indicate the starting and ending time for each of the activities below and provide remarks –
 

    
Activity Start End Remarks
Time Time
1
ix{d 
Download alert into PC in HIH (i.e. hih email inbox)
2 ix{dj ,enqKq njg ,ndoQka mqo.

a ,hdg oekajSu
Send acknowledgement to Alert Sender at source
3 
Create Event of Interest Document
4 
Obtain approval from Sarvodaya Executive
5 DEWNS RAD  CAP


Generate CAP text messages using DEWNS to send to
RAD
7 DEWNS  CAP


Generate CAP text message using DEWNS and
submit to Mobile Phones
222
8 DWRR (AREA) ANNY CAP


Generate CAP text message using ANNY to submit
DWRR (AREA)
9 CAP MP3 file  950 pek,h

fj; we;=,a lrkak
Generate CAP audio message as MP3 file and upload

on to channel 950
10 CAP  CDMA

.
Call and read CAP text message to CDMA users
11
IPAS Ndú;d lrñka CAP VSAT fj; Ndr
fokak
Generate CAP text message using IPAS and submit to
VSATs
12 .ñ jdiSkaf.ka iy uOHiaMdk u.ska ±kaùñ

,nd .kak
Recieve acknowledgements for the messages from
Villages and District Centres

Additional Information to be answered simply circling either ‘Yes’ or ‘No’ and aproviding additional remarks to justify

Remarks:
223
1
3 
 
 
Did the HIH practice the Standard procedures outlined in the “Guidelines for
Yes No
HIH”.
1 CAP hardware and software

4 ? 

Does the HIH have the necessay hardware and software to function as a CAP
Yes
message relay? No
1  
   
5 ?
Were the HIH staff competant in handling all situations that challenged them? Yes No

Additional Remarks (please use additional paper if necessary)
w;aik / ifnahg;gk; Èkh /jpfjp

Signature Date
224

Hazinfo Technical Report

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hazinfo Technical Report

Uploaded by

Copyright:

Available Formats

Evaluating Last-Mile Hazard Information

Dissemination: A Research Proposal

Grant No.: 103553-001

Grant Period: 01-December-2006 to 31-March-2008

Report Presentation Date: 10-March-2008

Submitted to IDRC by:

LIRNEasia is regional, non-

3.1.PERFORMANCE OF THE ICTS AS A WARNING TECHNOLOGY......................................................................................................18

4.FULLFILLMENT OF PROJECT OBJECTIVES..........................................................................32

5.PROJECT DESIGN AND IMPLEMENTATION..........................................................................36

6.PROJECT OUTPUT AND DISSEMINATION..............................................................................41

6.1.INFORMATION SHARING AND DISSEMINATIONS........................................................................................................................41

7.1.LOCAL RESEARCH AND DEVELOPMENT..................................................................................................................................48

10. OVERALL ASSESSMENT............................................................................................................52

11.1.MESSAGE FORMATTING - THE HIH AND THE GOVERNMENT....................................................................................................53

12.TECHNICAL ANNEX 1: LIST ACRONYMS AND ABBREVIATIONS.................................59

13.TECHNICAL ANNEX 2: WHAT IS A COMMUNITY-BASED LM-HWS..............................61

14.TECHNICAL ANNEX 3: INTERNETWORKING PIPES, AND GATEWAYS......................62

15.TECHNICAL ANNEX 4: RESEARCH MATRIX AND ICT ALLOCATIONS.......................66

16.TECHNICAL ANNEX 5: CALCULATING THE EFFICIENCY OF PROCESSES...............67

17.TECHNICAL ANNEX 6: MEASURING EFFECTIVENESS OF THE HIH AND

18.TECHNICAL ANNEX 7: CALCULATING RELIABILITY OF ICT.......................................70

18.1.CERTAINTY OF MESSAGE RECEIPT......................................................................................................................................71

19.TECHNICAL ANNEX 8: MEASURING EFFECTIVENESS OF THE ICT............................73

19.1.EFFECTIVENESS OF THE INDIVIDUAL ICTS............................................................................................................................73

20.TECHNICAL ANNEX 9: FUZZIFICATION OF FULL CAP COMPLETENESS..................74

20.1.CAP PROFILE FOR SRI LANKA..........................................................................................................................................74

21.TECHNICAL ANNEX 10: FUZZIFICATION OF ALERTING PROCESS.............................77

21.1.ACKNOWLEDGEMENT OF MESSAGE RECEIPT.........................................................................................................................77

22.TECHNICAL ANNEX 11: FUZZIFICATION OF TWO-WAY COMMUNICATION...........79

22.1.BI-DIRECTIONALITY FOR UPSTREAM AND DOWNSTREAM COMMUNICATION...............................................................................79

23.TECHNICAL ANNEX 12: FUZZIFICATION OF ADOPTABILITY.......................................80

23.1.INTEGRATION OF ICT IN TO COMMUNITY DAILY LIFE.............................................................................................................80

24.TECHNICAL ANNEX 13: FUZZIFICATION OF MINIATURERIZATION..........................82

24.1.WEIGHT OF THE ENTIRE FUNCTIONAL UNIT............................................................................................................................83

25.TECHNICAL ANNEX 14: EXPECTED PERFORMANCE OF THE ICT...............................86

25.1.WORLDSPACE ADDRESSABLE SATELLITE RADIO SYSTEM......................................................................................................86

26.TECHNICAL ANNEX 15: GUIDELINES FOR THE HIH.........................................................99

27.TECHNICAL ANNEX 16: TRAINING – HAZARDS...............................................................127

28.TECHNICAL ANNEX 17: TRAINING – VULNERABILITY AND RISK.............................137

29.TECHNICAL ANNEX 18: TRAINING – DISASTER MANAGEMENT................................145

29.1. DISASTER MANAGEMENT................................................................................................................................................145

30.TECHNICAL ANNEX 19: TRAINING – PARTICIPATORY RURAL APPRAISAL TOOLS

30.1.IMPORTANCE OF PARTICIPATORY TECHNIQUES....................................................................................................................150

31.TECHNICAL ANNEX 20: TRAINING – COMMUNITY PUBLIC AWARENESS..............155

31.1.DEFINITION OF SOCIAL CHANGE COMMUNICATION CAMPAIGN..............................................................................................155

32.TECHNICAL ANNEX 21: TRAINING – EARLY WARNING................................................162

32.1.CLASSIFICATION OF DISASTERS FOR EARLY WARNING ACTIVITIES............................................................................................163

33.TECHNICAL ANNEX 22: TRAINING – RESPONSE PLANNING.......................................166

34.TECHNICAL ANNEX 23: TRAINING –MONITORING AND EVALUATION...................171

34.1.DEFINITION OF MONITORING AND EVALUATION..................................................................................................................171

35.APPENDIX A: MAP OF SRI LANKA WITH 32 COASTAL COMMUNITIES....................177

36.APPENDIX B: DATA OF RESEARCH MATRIX ALLOCATION........................................178

37.APPENDIX C: DATA OF HIH OPERTIONAL.........................................................................179

38.APPENDIX D: DATA OF EFFICIENCY MEASURE FROM LIVE-EXERCISES..............182

39.APPENDIX E: DATA FOR CERTAINTY MEAURE OF ICT................................................187

40.APPENDIX F: EFFECTIVENESS DATA FOR ALL THE CLIQUES...................................190

40.1.FULL CAP COMPLETENESS.............................................................................................................................................190

41.APPENDIX G: OUTCOME OF THE ICT IN THE FIELD......................................................195

41.1.WORLDSPACE ADDRESSABLE SATELLITE RADIO SYSTEM....................................................................................................195

42.APPENDIX H – DATA FROM SILENT TESTS........................................................................202

43.APPENDIX I – OUTCOME OF ICT ON 12-SEP-2007 TSUNAMI ADVISORY...................209

44.APPENDIX J – DATA COLLECTION SHEETS......................................................................211

44.1.DATASHEET 1 – ICT GUARDIAN ALERT LOG....................................................................................................................211